The present grievous situation of the tuberculosis disease can be improved by efficient case management and timely follow-up evaluations. With the advent of digital technology this can be achieved by quick summarization of the patient-centric data. The aim of our study was to assess the effectiveness of the ODK Scan paper-to-digital system during testing period of three months. A sequential, explanatory mixed-method research approach was employed to elucidate technology use. Training, smartphones, application and 3G enabled SIMs were provided to the four field workers. At the beginning, baseline measures of the data management aspects were recorded and compared with endline measures to see the impact of ODK Scan. Additionally, at the end, users’ feedback was collected regarding app usability, user interface design and workflow changes. 122 patients’ records were retrieved from the server and analysed for quality. It was found that ODK Scan recognized 99.2% of multiple choice bubble responses and 79.4% of numerical digit responses correctly. However, the overall quality of the digital data was decreased in comparison to manually entered data. Using ODK Scan, a significant time reduction is observed in data aggregation and data transfer activities, however, data verification and form filling activities took more time. Interviews revealed that field workers saw value in using ODK Scan, however, they were more concerned about the time consuming aspects of the use of ODK Scan. Therefore, it is concluded that minimal disturbance in the existing workflow, continuous feedback and value additions are the important considerations for the implementing organization to ensure technology adoption and workflow improvements.

Educational data-mining is an evolving discipline that focuses on the improvement of self-learning and adaptive methods. It is used for finding hidden patterns or intrinsic structures of educational data. In the arena of education, the heterogeneous data is involved and continuously growing in the paradigm of big-data. To extract meaningful information adaptively from big educational data, some specific data mining techniques are needed. This paper presents a clustering approach to partition students into different groups or clusters based on their learning behavior. Furthermore, personalized e-learning system architecture is also presented which detects and responds teaching contents according to the students’ learning capabilities. The primary objective includes the discovery of optimal settings, in which learners can improve their learning capabilities. Moreover, the administration can find essential hidden patterns to bring the effective reforms in the existing system. The clustering methods K-Means, K-Medoids, Density-based Spatial Clustering of Applications with Noise, Agglomerative Hierarchical Cluster Tree and Clustering by Fast Search and Finding of Density Peaks via Heat Diffusion (CFSFDP-HD) are analyzed using educational data mining. It is observed that more robust results can be achieved by the replacement of existing methods with CFSFDP-HD. The data mining techniques are equally effective to analyze the big data to make education systems vigorous.

The importance of energy systems and its role in economics and politics is not hidden for anyone. This issue is not only important for the advanced industrialized countries, which are major energy consumers, but is also important for oil-rich countries. In addition to the nature of these fuels which contains polluting substances, the issue of their ending up has aggravated the growing concern. Biofuels can be used in different fields for energy production like electricity production, power production or for transportation. Various scenarios have been written about the estimated biofuels from different sources in the future energy system. The availability of biofuels for the electricity market, heating and liquid fuels is very important. Accordingly, the need for handling, modelling, decision making and future forecasting for biofuels can be one of the main challenges for scientists. Recently, machine learning and deep learning techniques have been popular in modeling, optimizing and handling the biodiesel production, consumption and its environmental impacts. The main aim of this study is to evaluate the ML and DL techniques developed for handling biofuels production, consumption and environmental impacts, both for modeling and optimization purposes. This will help for sustainable biofuel production for the future generations.

As Artificial Intelligence (AI) technology, particularly Machine Learning algorithms, becomes increasingly ubiquitous, our abilities to understand and interpret AI and machine learning algorithms are becoming increasingly desirable. Visualization has been used as a common tool to view and understand complex machine learning processes. In this paper, we will focus on developing a general framework for the visualization of machine learning models as a scalar valued multi-dimensional function to help users understand how the models behave over different viewing spaces. We first give a formal definition of the visualization problem. Interpolation-based morphing and subspace sampling techniques are applied to generate various renderings through projections and cross-sections of the model space as 3D surfaces or heatmap images. This method will be applied to two real-world datasets and applications: the diagnosis of Alzheimer's Disease (AD) using a human brain networks dataset and a real-world benchmark dataset for predicting home credit default risks. The visualizations show that different machine learning algorithms can behave quite differently under different conditions.

CVDs are a leading cause of death globally. In CVDs, the heart is unable to deliver enough blood to other body regions. Since effective and accurate diagnosis of CVDs is essential for CVD prevention and treatment, machine learning (ML) techniques can be effectively and reliably used to discern patients suffering from a CVD from those who do not suffer from any heart condition. Namely, machine learning algorithms (MLAs) play a key role in the diagnosis of CVDs through predictive models that allow us to identify the main risks factors influencing CVD development. In this study, we analyze the performance of ten MLAs on two datasets for CVD prediction and two for CVD diagnosis. Algorithm performance is analyzed on top-two and top-four dataset attributes/features with respect to five performance metrics –accuracy, precision, recall, f1-score, and roc-auc – using the train-test split technique and k-fold cross-validation. Our study identifies the top two and four attributes from each CVD diagnosis/prediction dataset. As our main findings, the ten MLAs exhibited appropriate diagnosis and predictive performance; hence, they can be successfully implemented for improving current CVD diagnosis efforts and help patients around the world, especially in regions where medical staff is lacking.

Disease surveillance is critical to monitor ongoing control activities, detect early outbreaks and to inform intervention priorities and policies. Unfortunately, most data from disease surveillance remain under-utilised to support decision-making in real-time. Using the Brazilian Amazon malaria surveillance data as a case study, we explore unsupervised anomaly detection machine learning techniques to analyse and discover potential anomalies. We found that our models are able to detect early outbreaks, peak of outbreaks as well as change points in the proportion of positive malaria cases. Specifically, the sustained rise in malaria in the Brazilian Amazon in 2016 was flagged by several models. We also found that no single model detects all the anomalies across all health regions. The approaches using Clustering-based local outlier algorithm ranked first before Principal component analysis and Stochastic outlier selection in maximising the number of anomalies detected in local health regions. Because of this, we also provide the minimum number of machine learning models (top-k models) to maximise the number of anomalies detected across different health regions. We discovered that the top-3 models that maximise the coverage of the number and types of anomalies detected across the 13 health regions are: Principal component analysis, Stochastic outlier selection and Multi-covariance determinant. Anomaly detection approaches provide interesting solutions to discover patterns of epidemiological importance when confronted with a large volume of data across space and time. Our exploratory approach can be replicated for other diseases and locations to inform timely interventions and actions toward endemic disease control.

Biofuels construct an essential pillar of energy systems. Biofuels are considered as a popular resource for electricity production, heating, household, and industrial usage, liquid fuels, and mobility around the world. Thus, the need for handling, modeling, decision-making, demand, and forecasting for biofuels are of utmost importance. Recently, machine learning (ML) and deep learning (DL) techniques have been accessible in modeling, optimizing, and handling biofuels production, consumption, and environmental impacts. The main aim of this study is to review and evaluate ML and DL techniques and their applications in handling biofuels production, consumption, and environmental impacts, both for modeling and optimization purposes. Hybrid and ensemble ML methods, as well as DL methods, have found to provide higher performance and accuracy in modeling the biofuels.

As data driven models gain importance in driving decisions and processes, recently it becomes increasingly important to visualize the data with both speed and accuracy. A massive amount of data is presently generated in the educational sphere from various learning platforms, tools, and institutions. Visual analytics of educational big data has capability to improve student learning, develop strategies for personalized learning, and improve faculty productivity. How-ever, there are limited advancements in the education domain for data-driven decision making leveraging the recent advancements in the field of machine learning. Some of the recent tools such as Tableau, Power BI, Microsoft Azure suite, Sisense etc. leverage artificial intelligence and machine learning techniques to visualize data and generate insights out of it, however their ap-plicability in educational advances is limited. This paper focuses on leveraging machine learn-ing and visualization techniques to demonstrate their utility through a practical implementation using k-12 state assessment data compiled from the institutional websites of the State of Texas and Louisiana. Effective modeling and predictive analytics are the focus of the sample use case presented in this research. Our approach demonstrates the applicability of web technology in conjunction with machine learning to provide a cost-effective and timely solution to visualize and analyze big educational data. Additionally, ad-hoc visualization provides contextual anal-ysis within areas of concern for Education Agencies (EA).

Dyadic data which is also called co-occurrence data (COD) contains co-occurrences of objects. Searching for statistical models to represent dyadic data is necessary. Fortunately, finite mixture model is a solid statistical model to learn and make inference on dyadic data because mixture model is built smoothly and reliably by expectation maximization (EM) algorithm which is suitable to inherent spareness of dyadic data. This research summarizes mixture models for dyadic data. When each co-occurrence in dyadic data is associated with a value, there are many unaccomplished values because a lot of co-occurrences are inexistent. In this research, these unaccomplished values are estimated as mean (expectation) of random variable given partial probabilistic distributions inside dyadic mixture model.

The work presented in this paper makes multiple scientific contributions related to the investigation of the global fear associated with COVID-19 by performing a comprehensive analysis of a dataset comprising survey responses of participants from 40 countries. First, the results of subjectivity analysis of responses where participants indicated their biggest concern related to COVID-19 showed that the average subjectivity in responses by the age group of 41-50 decreased from April 2020 to June 2020, the average subjectivity in responses by the age group of 71-80 drastically increased from May 2020, and the age group of 11-20 indicated the least level of subjectivity in their responses between June 2020 to August 2020. Second, subjectivity analysis also revealed the percentage of highly opinionated, neutral opinionated, and least opinionated responses per age-group where the analyzed age groups were 11-20, 21-30, 31-40, 41-50, 51-60, 61-70, 71-80, and 81-90. For instance, the percentage of highly opinionated, neutral opinionated, and least opinionated responses by the age group of 11-20 were 17.92%, 16.24%, and 65.84%, respectively. Third, data analysis of responses from different age groups showed that the highest percentage of responses indicating that they were very worried about COVID-19 came from individuals in the age group of 21-30. Fourth, data analysis of the survey responses also revealed that in the context of taking precautions to prevent contracting COVID-19, the percentage of individuals in the age group of 31-40 taking precautions was higher as compared to the percentages of individuals from the age groups of 41-50, 51-60, 61-70, 71-80, and 81-90. Finally, a deep learning model was developed to detect if the survey respondents were seeing or planning to see a psychologist or psychiatrist for any mental health issues related to COVID-19. The deep learning model used the responses to multiple questions in the context of fear, preparedness, and response related to COVID-19 from the dataset and achieved an overall performance accuracy of 91.62% after 500 epochs.

Deep learning (DL) and machine learning (ML) methods have recently contributed to the advancement of models in the various aspects of prediction, planning, and uncertainty analysis of smart cities and urban development. This paper presents the state of the art of DL and ML methods used in this realm. Through a novel taxonomy, the advances in model development and new application domains in urban sustainability and smart cities are presented. Findings reveal that five DL and ML methods have been most applied to address the different aspects of smart cities. These are artificial neural networks; support vector machines; decision trees; ensembles, Bayesians, hybrids, and neuro-fuzzy; and deep learning. It is also disclosed that energy, health, and urban transport are the main domains of smart cities that DL and ML methods contributed in to address their problems.

The effects of the 2008 financial crisis undoubtedly caused problems not only to the banking sector but also to the real economy of the developed and the developing countries in almost all around the globe. Besides, as is widely known, every banking crisis entails the corresponding cost to the economy of each country affected by it, which results from the shakeout and the restructuring of its financial system. The purpose of this research is to investigate the consequences of the financial crisis and the COVID-19 health crisis and how these affected the course of the four systemic banks (Eurobank, Alpha Bank, National Bank, Piraeus Bank) through the analysis of ratios for the period of 2015-2020.

In the current age of the Fourth Industrial Revolution ($4IR$ or Industry $4.0$), the digital world has a wealth of data, such as Internet of Things (IoT) data, cybersecurity data, mobile data, business data, social media data, health data, etc. To intelligently analyze these data and develop the corresponding real-world applications, the knowledge of artificial intelligence (AI), particularly, machine learning (ML) is the key. Various types of machine learning algorithms such as supervised, unsupervised, semi-supervised, and reinforcement learning exist in the area. Besides, the deep learning, which is part of a broader family of machine learning methods, can intelligently analyze the data on a large scale. In this paper, we present a comprehensive view on these machine learning algorithms that can be applied to enhance the intelligence and the capabilities of an application. Thus, this study's key contribution is explaining the principles of different machine learning techniques and their applicability in various real-world applications areas, such as cybersecurity, smart cities, healthcare, business, agriculture, and many more. We also highlight the challenges and potential research directions based on our study. Overall, this paper aims to serve as a reference point for not only the application developers but also the decision-makers and researchers in various real-world application areas, particularly from the technical point of view.

The concept of ‘big data’ has been widely discussed, and its value has been illuminated throughout a variety of domains. To quickly mine potential values and alleviate the ever-increasing volume of information, machine learning is playing an increasingly important role and faces more challenges than ever. Because few studies exist regarding how to modify machine learning techniques to accommodate big data environments, we provide a comprehensive overview of the history of the evolution of big data, the foundations of machine learning, and the bottlenecks and trends of machine learning in the big data era. More specifically, based on learning principals, we discuss regularization to enhance generalization. The challenges of quality in big data are reduced to the curse of dimensionality, class imbalances, concept drift and label noise, and the underlying reasons and mainstream methodologies to address these challenges are introduced. Learning model development has been driven by domain specifics, dataset complexities, and the presence or absence of human involvement. In this paper, we propose a robust learning paradigm by aggregating the aforementioned factors. Over the next few decades, we believe that these perspectives will lead to novel ideas and encourage more studies aimed at incorporating knowledge and establishing data-driven learning systems that involve both data quality considerations and human interactions.

A heterogeneous disease like cancer is activated through multiple pathways and different perturbations. Depending upon the activated pathway(s), patients’ survival vary significantly and show different efficacy to various drugs. Therefore, cancer subtype detection using genomics level data is a significant research problem. Subtype detection is often a complex problem, and in most cases, needs multi-omics data fusion to achieve accurate subtyping. Different data fusion and subtyping approaches have been proposed, such as kernel-based fusion, matrix factorization, and deep learning autoencoders. In this paper, we compared the performance of different deep learning autoencoders for cancer subtype detection. We performed cancer subtype detection on four different cancer types from The Cancer Genome Atlas (TCGA) datasets using four autoencoder implementations. We also predicted the optimal number of subtypes in a cancer type using the silhouette score. We observed that the detected subtypes exhibit significant differences in survival profiles. Furthermore, we also compared the effect of feature selection and similarity measures for subtype detection. To evaluate the results obtained, we selected the Glioblastoma multiforme (GBM) dataset and identified the differentially expressed genes in each of the subtypes identified by the autoencoders; the obtained results coincide well with other genomic studies and can be corroborated with the involved pathways and biological functions. Thus, it shows that the results from the autoencoders, obtained through the interaction of different datatypes of cancer, can be used for the prediction and characterization of patient subgroups and survival profiles.

This paper provides the state of the art of data science in economics. Through a novel taxonomy of applications and methods advances in data science are investigated. The data science advances are investigated in three individual classes of deep learning models, ensemble models, and hybrid models. Application domains include stock market, marketing, E-commerce, corporate banking, and cryptocurrency. Prisma method, a systematic literature review methodology is used to ensure the quality of the survey. The findings revealed that the trends are on advancement of hybrid models as more than 51% of the reviewed articles applied hybrid model. On the other hand, it is found that based on the RMSE accuracy metric, hybrid models had higher prediction accuracy than other algorithms. While it is expected the trends go toward the advancements of deep learning models.

The copper mining industry is increasingly using artificial intelligence methods to improve cop-per production processes. Recent studies reveal the use of algorithms such as Artificial Neural Network, Support Vector Machine, and Random Forest, among others, to develop models for predicting product quality. Other studies compare the predictive models developed with these machine learning algorithms in the mining industry, as a whole. However, not many copper mining studies published compare the results of machine learning techniques for copper recovery prediction. This study makes a detailed comparison between three models for predicting copper recovery by leaching, using four datasets resulting from mining operations in northern Chile. The algorithms used for developing the models were Random Forest, Support Vector Machine, and Artificial Neural Network. To validate these models, four indicators or values of merit were used: accuracy (acc), precision (p), recall (r), and Matthew’s correlation coefficient (mcc). This paper describes dataset preparation and the refinement of the threshold values used for the predictive variable most influential on the class (the copper recovery). Results show both a precision over 98.50% and also the model with the best behavior between the predicted and the real. Finally, the models obtained show the following mean values: acc=94.32, p=88.47, r=99.59, and mcc=2.31. These values are highly competitive as compared with those obtained in similar studies using other approaches in the context.

Prediction of protein-ligand binding affinities is crucial for computational drug discovery. A number of deep learning approaches have been developed in recent years to improve the accuracy of such affinity prediction. While the predicting power of these systems have advanced to some degrees depending on the dataset used for model training and testing, the effects of the quality and quantity of the underlying data have not been thoroughly examined. In this study, we employed erroneous datasets and data subsets of different sizes, created from one of the largest databases of experimental binding affinities, to train and evaluate a deep learning system based on convolutional neural networks. Our results show that data quality and quantity do have significant impacts on the prediction performance of trained models. Depending on the variations in data quality and quantity, the performance discrepancies could be comparable to or even larger than those observed among different deep learning approaches. In particular, the presence of proteins during model training leads to a dramatic increase in prediction accuracy. This implies that continued accumulation of high-quality affinity data, especially for new protein targets, is indispensable for improving deep learning models to better predict protein-ligand binding affinities.

Data mining techniques have garnered significant attention within the realm of education. However, the procurement of ample student data poses a formidable challenge. In response to this challenge, we present a student dataset characterized by its size and distinctive attributes. This dataset encompasses various task-related topics interconnected through a learning pathway, thereby enabling researchers to delve into the data from novel perspectives. Moreover, it encompasses extensive longitudinal student behavioral data, a rarity that adds substantial value. Spanning the years from 2010 to 2021, our dataset comprises a cohort of 7,933 students, 64,344 test scores, and 183,390 behavior records, solidifying its status as a valuable resource for educational research. In our experiments, we achieved successful predictions of students' test outcomes based on behavioral learning data. The strengths of our dataset render it apt for analyzing the nexus between student conduct and academic performance, crafting personalized learning recommendations, and pursuing various other research pursuits.

The monitoring of surface-water quality followed by water-quality modeling and analysis is essential for generating effective strategies in water-resource management. However, worldwide, particularly in developing countries, water-quality studies are limited due to the lack of a complete and reliable dataset of surface-water-quality variables. In this context, several statistical and machine-learning models were assessed for imputing water-quality data at six monitoring stations located in the Santa Lucía Chico river (Uruguay), a mixed lotic and lentic river system. The challenge of this study is represented by the high percentage of missing data (between 50% and 70%) and the high temporal and spatial variability that characterizes the water-quality variables. The competing algorithms implemented belonged to both univariate and multivariate imputation methods (inverse distance weighting (IDW), Random Forest Regressor (RFR), Ridge (R), Bayesian Ridge (BR), AdaBoost (AB), Hubber Regressor (HR), Support Vector Regressor (SVR), and K-nearest neighbors Regressor (KNNR)). According to the results, more than 76% of the imputation outcomes are considered satisfactory (NSE > 0.45). The imputation performance shows better results at the monitoring stations located inside the reservoir than the ones positioned along the mainstream. IDW was the most chosen model for data imputation.

This paper presents a study on the thermal efficiency improvement of the user equipment room in the district heating system based on reinforcement learning [1], and suggests a general method of constructing a learning network(DQN)[2] using deep Q learning[3], which is a reinforcement learning algorithm that does not specify a model. In addition, we introduce the big data platform system and the integrated heat management system for the energy field in the massive data processing from the IoT sensor installed in large number of thermal energy control facilities.

Biofilm is a complex community of microorganisms that are attached to surfaces and encased in a self-produced extracellular matrix. Machine learning (ML) techniques have been applied to various aspects of biofilm research, such as predicting biofilm formation, identifying key genes, and designing new therapeutic strategies. In this study, we conducted a bibliometric analysis of machine learning in biofilm research to provide a comprehensive overview of the current state of the field. We searched the Web of Science database for articles published included "machine learning biofilm". A total of 126 articles were identified and analysed. Our results showed that the number of publications on machine learning in biofilm has been increasing rapidly over the past decade, indicating a growing interest in the application of ML techniques to biofilm research. The analysis also revealed that the most common research topics in this area were related to biofilm formation, prediction, and control. Furthermore, the most frequently used ML techniques in biofilm research were artificial neural networks and support vector machines. Overall, our study provides valuable insights into the current trends and future directions of machine learning in biofilm research. It also highlights the importance of interdisciplinary collaboration between biofilm researchers and ML experts to drive innovation in this field.

This paper explores the application of various machine learning techniques for predicting customer churn in the telecommunications sector. We utilized a publicly accessible dataset and implemented several models, including Artificial Neural Networks, Decision Trees, Support Vector Machines, Random Forests, Logistic Regression, and gradient boosting techniques (XGBoost, LightGBM, and CatBoost). To mitigate the challenges posed by imbalanced datasets, we adopted different data sampling strategies, namely SMOTE, SMOTE combined with Tomek Links, and SMOTE combined with Edited Nearest Neighbors. Moreover, hyperparameter tuning was employed to enhance model performance. Our evaluation employed standard metrics such as Precision, Recall, F1-Score, and the Receiver Operating Characteristic Area Under Curve (ROC AUC). Regarding the F1-Score metric, CatBoost demonstrates superior performance compared to other machine learning models, achieving an outstanding 93% following the application of Optuna hyperparameter optimization. In the context of the ROC AUC metric, both XGBoost and CatBoost exhibit exceptional performance, recording remarkable scores of 91%. This achievement for XGBoost is attained after implementing a combination of SMOTE with Tomek Links, while CatBoost reaches this level of performance after the application of Optuna hyperparameter optimization.

Object classification is a crucial task in deep learning, which involves the identification and categorization of objects in images or videos. Although humans can easily recognize common objects, such as cars, animals, or plants, performing this task on a large scale can be time-consuming and error-prone. Therefore, automating this process using neural networks can save time and effort while achieving higher accuracy. Our study focuses on the classification step of human chromosome karyotyping, an important medical procedure that helps diagnose genetic disorders. Traditionally, this task is performed manually by expert cytologists, which is a time-consuming process that requires specialized medical skills. Therefore, automating it through deep learning can be immensely useful. To accomplish this, we implemented and adapted existing preprocessing and data augmentation techniques to prepare the chromosome images for classification. We used ResNet-50 convolutional neural networks and an ensemble approach to classify the chromosomes, obtaining state-of-the-art performance in the tested dataset.

Scientists and astronomers have attached Scientists and astronomers have attached great importance to the task of discovering new exoplanets, even more so if they are in the habitable zone. To date, more than 4300 exoplanets have been confirmed by NASA, using various discovery techniques, including planetary transits, in addition to the use of various databases provided by space and ground-based telescopes. This article proposes the development of a deep learning system for detecting planetary transits in Kepler Telescope lightcurves. The approach is based on related work from the literature and enhanced to validation with real lightcurves. A CNN classification model is trained from a mixture of real and synthetic data, and validated only with real data and different from those used in the training stage. The best ratio of synthetic data is determined by the perform of an optimisation technique and a sensitivity analysis. The precision, accuracy and true positive rate of the best model obtained are determined and compared with other similar works. The results demonstrate that the use of synthetic data on the training stage can improve the transit detection performance on real light curves.

Artificial Intelligence (AI) models are expected to have a great impact in the manufacturing industry, optimizing time and resource cost by enabling applications such as predictive maintenance (PM) of production machines. A necessary condition for this is the availability of high quality data collected as close as possible to the process in question. With the advent of 5G equipped multi sensor platforms (MSPs), high sampling rate data can be collected and transmitted for processing in real time. This poses a data security challenge, since this data may give valuable insight into confidential business information of companies. Federated learning (FL) enables the training of AI models with data from multiple sources without it leaving the shop floor, by utilizing distributed computing resources available on premise. This paper introduces an architecture of FL based on data collected from 5G MSPs for enabling PM in industrial environments and discusses its potential benefits and challenges.

Although learning analytics (LA) holds great potential to improve teaching and learning, LA research and practice are currently riddled with limitations that impact every stage of the LA life cycle. The present paper offers an overview of these challenges before proposing strategies to overcome them and exploring how the recent innovations brought forth by language models can improve LA research and practice. In particular, we encourage the empowerment of teachers during LA development, as this would strengthen the theoretical foundation of LA solutions and increase their interpretability and usability. Furthermore, we provide examples of how process data can be used to understand learning processes and generate more interpretable LA insights. Furthermore, we explore how LLMs could come into play in LA to generate interpretable insights, timely and actionable feedback, increase personalization, and support teachers’ tasks more broadly.

Sparse reward long horizon task is a major challenge for deep reinforcement learning algorithm. One of the key barriers is data-inefficiency. Even in the simulation environment, it usually takes weeks to training the agent. In this study, a data-efficiency training framework is proposed, where a curriculum learning is design for the agent in the simulation scenario. Different distributions of the initial state are set for the agent to get more informative reward during the whole training process. A fine-tuning of the parameters in the output layer of the neural network for value function is conduct to bridge the gap between sim-to-real. An experiment of UAV maneuver control is conducted in the proposed training framework to verify the method more efficient. We demonstrate that data-efficiency is different for the same data in different training stages.

This study introduces the Dynamic Feedback-Driven Learning Optimization Framework (DFDLOF), an innovative machine learning-based approach to personalize educational pathways. DFDLOF leverages advanced algorithms for in-depth analysis of learner data, enabling the dynamic customization of educational content and methodologies. Its application in diverse online platforms, including Khan Academy and Coursera, underscores its adaptability and scalability. DFDLOF enhances learner engagement and outcomes by providing tailored learning experiences and real-time feedback. Key insights reveal its strengths in personalized education, particularly in dynamically adapting learning paths and the importance of real-time feedback. Challenges identified include integration complexities and data privacy concerns. Future research directions emphasize the need for longitudinal impact studies and broader demographic applications. This study contributes significantly to the evolving domain of adaptive learning technologies, demonstrating the transformative potential of machine learning in personalizing education.

This paper proposes a Grouped Contrastive Learning of self-supervised Sentence Representation (GCLSR), which can learn an effective and meaningful representation of sentences. Previous works maximize the similarity between two vectors to be the objective of contrastive learning, suffering from the high-dimensionality of the vectors. In addition, most previous works have adopted discrete data augmentation to obtain positive samples and directly employed contrastive framework of computer vision to perform contrastive training, which could hamper contrastive training because text data is discrete and sparse compared with image data. To address those issues, we propose a grouped contrastive learning framework, i.e., GCLSR, which divides the high-dimensional feature vector into several groups and respectively computes the groups’ contrastive losses to make use of more local information, eventually obtaining a more fine-grained sentence representation. In addition, in GCLSR, we design a new self-attention mechanism and a continuous as well as partial word vector augmentation (PWVA). For the discrete and sparse text data, the usage of self-attention could help model focus the informative words by measuring the importance of every word in a sentence. By using the PWVA, GCLSR can obtain high-quality positive samples used for contrastive learning. Experimental results demonstrate that our proposed GCLSR achieves an encouraging result on the challenging datasets of the standard semantic textual similarity (STS) task and transfer task.

Abstract:Global ship demolition is mostly concentrated in south Asian countries, namely Bangladesh, India, Pakistan and China, since 1990’s, having competitive advantage for their high natural tide, and low environmental and social costs. Due to high social and environmental externalities, stakeholders increase monitoring of the externalities and continue to prescribe improvement towards sustainability, which put pressures on profitability and competitiveness. As a consequence, also seen in the past, a leakage effect may emerge, leading to shift of this activity to a region, with relatively less monitored and less stricter on social and environmental impacts. Unfortunately, the leakage effect is never predicted in shipbreaking in order to understand the level of push compatible in the given socio-economic contexts. In this study, we have attempted to predict the future ship demolition landscape, applying machine learning technique to 34,531 in-service vessels worldwide, larger than 500 gross tonnage (GT), which is run against a learning model based on 3500 demolished vessels from 2014. This study shows that redistribution may occur among the top recycling nations: India may emerge out to be a dominant player in shipbreaking, surpassing Bangladesh by a margin of two-fold, while Pakistan and China are in decreasing trend. In addition, the leakage effect is observed, in that Vietnam is predicted to be the fourth largest ship demolition country, while China and Pakistan recede from the third and fourth place to 6th and 8th. Turkey is predicted to advance from fifth position to third position by vessel count but stays same in term of total GT dismantled. Although it is not clear if any leakage is to be observed in the near future, this study may be a model for future predictive analytics and help stakeholders take evidence-based business decisions.

Now when the whole world is still under COVID-19 pandemic, many schools have transferred the teaching from physical classroom to online platforms. It is highly important for schools and online learning platforms to investigate the feedback to get valuable insights about online teaching process so that both platforms and teachers are able to learn which aspect they can improve to achieve better teaching performance. But handling reviews expressed by students would be a pretty laborious work if they were handled manually as well as it is unrealistic to handle large-scale feedback from e-learning platform. In order to address this problem, both machine learning algorithms and deep learning models are used in recent research to automatically process students' review getting the opinion, sentiment and attitudes expressed by the students. Such studies may play a crucial role in improving various interactive online learning platforms by incorporating automatic analysis of feedback. Therefore, we conduct an overview study of sentiment analysis in educational field presented in recent research, to help people grasp an overall understanding of the sentiment analysis research. Besides, according to the literature review, we identify three future directions that researchers can focus on in automatically feedback processing: high-level entity extraction, multi-lingual sentiment analysis, and handling of figurative language.

Emotion detection from the text is an important and challenging problem in text analytics. The opinion-mining experts are focusing on the development of emotion detection applications as they have received considerable attention of online community including users and business organization for collecting and interpreting public emotions. However, most of the existing works on emotion detection used less efficient machine learning classifiers with limited datasets, resulting in performance degradation. To overcome this issue, this work aims at the evaluation of the performance of different machine learning classifiers on a benchmark emotion dataset. The experimental results show the performance of different machine learning classifiers in terms of different evaluation metrics like precision, recall ad f-measure. Finally, a classifier with the best performance is recommended for the emotion classification.

As the development of high-throughput technologies, more and more high-dimensional or ultra high-dimensional genomic data are generated. Therefore, how to make effective analysis of such data becomes a challenge. Machine learning (ML) algorithms have been widely applied for modelling nonlinear and complicated interactions in a variety of practical fields such as high-dimensional survival data. Recently, the multilayer deep neural network (DNN) models have made remarkable achievements. Thus, a Cox-based DNN prediction survival model (DNNSurv model) , which was built with Keras and Tensorflow, was developed. However, its results were only evaluated to the survival datasets with high-dimensional or large sample sizes. In this paper, we evaluate the prediction performance of the DNNSurv model using ultra high-dimensional and high-dimensional survival datasets, and compare it with three popular ML survival prediction models (i.e., random survival forest and Cox-based LASSO and Ridge models). For this purpose we also present the optimal setting of several hyper-parameters including selection of tuning parameter. The proposed method demonstrates via data analysis that the DNNSurv model performs overall well as compared with the ML models, in terms of three main evaluation measures (i.e., concordance index, time-dependent Brier score and time-dependent AUC) for survival prediction performance.

Digital Twin Cities (DTCs) play a fundamental role in city planning and management. They allow three-dimensional modeling and simulation of cities. 3D semantic segmentation is the foundation for automatically creating enriched DTCs, as well as their updates. Past studies indicate that prior level fusion approaches demonstrate more promising precisions in 3D semantic segmentation compared to point level fusion, features level fusion, and decision level fusion families. In order to improve point cloud enriched semantic segmentation outcomes, this article proposes a new approach for 3D point cloud semantic segmentation through developing and benchmarking three prior level fusion scenarios. A reference approach based on point clouds and aerial images was proposed to compare it with the different developed scenarios. In each scenario, we inject a specific prior knowledge (geometric features,classified images ,etc) and aerial images as attributes of point clouds into the neural network’s learning pipeline. The objective is to find the one that integrates the most significant prior knowledge and enhances neural network knowledge more profoundly, which we have named the "smart fusion approach". The advanced Deep Learning algorithm "RandLaNet" was adopted to implement the different proposed scenarios and the reference approach, due to its excellent performance demonstrated in the literature. The introduction of some significant features associated with the label classes facilitated the learning process and improved the semantic segmentation results that can be achievable with the same neural network alone. Overall, our contribution provides a promising solution for addressing some challenges, in particular more accurate extraction of semantically rich objects from the urban fabric. An assessment of the semantic segmentation results obtained by the different scenarios is performed based on metrics computation and visual investigations. Finally,the smart fusion approach was derived based on the obtained qualitative and quantitative results.

The prediction of system degradation is very important as it serves as an important basis for the formulation of condition-based maintenance strategies. An effective health indicator (HI) plays a key role in the prediction of system degradation as it enables vital information for critical tasks ranging from fault diagnosis to remaining useful life prediction. To address this issue, a method for monitoring data fusion and health indicator construction based on autoencoder (AE) and long short-term memory (LSTM) network is proposed to improve the predictability and effectiveness of health indicator in this study. Firstly, an unsupervised method and overall framework for HI construction is built based on deep autoencoder and LSTM neural network. The neural network is trained fully based on the normal operating monitoring data and then the construction error of the AE model is adopted as the health indicator of the system. Secondly, we propose related machine learning techniques for monitoring data processing to overcome the issue of data fusion, like mutual information for sensor selection and t-distributed stochastic neighbor embedding (T-SNE) for operating condition identification, etc. Thirdly, in order to verify the performance of the proposed method, experiments are conducted based on CMAPSS dataset and results are compared with algorithms of principal component analysis (PCA) and vanilla autoencoder model. Result shows that LSTM-AE model outperforms the PCA and Vanilla-AE model in metrics of monotonicity, trendability, prognosability and fitness. Fourthly, in order to analyze the impact of the time step of the LSMT-AE model on HI construction, we construct and analyze the system HI curve under different time steps of 5, 10, 15, 20, and 25 cycles. Finally, the results demonstrate that the proposed method for HI construction can effectively characterize the health state of system, which is helpful for further failure prognostics and converting the scheduled maintenance into condition-based maintenance.

Deep learning (DL) models have been recently widely used to extract task-oriented patterns from large scale of datasets, and to improve the data image understanding and analysis accuracy in many different decision-making processes for tasks such as image classification, segmentation, detection, and so on. However, in practice, the performances of DL models are easily affected by environmental illumination conditions. Conversely, DL models can also be utilized for extracting the illumination hints from the images, and these hints are critically useful for improving the model robustness, classifying the environmental scenes, estimating scene depth information, and rendering 3D objects. In this study, an extensive and exhaustive review is carried out for DL based color constancy, indoor and outdoor illumination estimation, and image depth estimations with the considerations of strengths and weaknesses of DL models. This study also explores the different network designs and the paradoxes in parameter optimization during the model training. Current technology barriers involved in implementing these models and recommendations to overcome these barriers are also suggested in the review.

In the field of computer vision, large-scale image classification tasks are both important and highly challenging. With the ongoing advances in deep learning and optical character recognition (OCR) technologies, neural networks designed to perform large-scale classification play an essential role in facilitating OCR systems. In this study, we developed an automatic OCR system designed to identify up to 13,070 large-scale printed Chinese characters by using deep learning neural networks and fine-tuning techniques. The proposed framework comprises four components, including training dataset synthesis and background simulation, image preprocessing and data augmentation, the process of training the model, and transfer learning. The training data synthesis procedure is composed of a character font generation step and a background simulation process. Three background models are proposed to simulate the factors of the background noise and anti-counterfeiting patterns on ID cards. To expand the diversity of the synthesized training dataset, rotation and zooming data augmentation are applied. A massive dataset comprising more than 19.6 million images was thus created to accommodate the variations in the input images and improve the learning capacity of the CNN model. Subsequently, we modified the GoogLeNet neural architecture by replacing the FC layer with a global average pooling layer to avoid overfitting caused by a massive amount of training data. Consequently, the number of model parameters was reduced. Finally, we employed the transfer learning technique to further refine the CNN model using a small number of real data samples. Experimental results show that the overall recognition performance of the proposed approach is significantly better than that of prior methods and thus demonstrate the effectiveness of proposed framework, which exhibited a recognition accuracy as high as 99.39% on the constructed real ID card dataset.

AliAmbra is a project developed to explore and promote high-quality catches of the Amvrakikos Gulf GP to Artas’ wider regions. In addition, this project aimed to implement an integrated plan of action, to form a business identity with high-added value and achieve integrated business services adapted to the special characteristics of the area. The action plan for this project was to actively search for new markets, create a collective identity for the products, promote their quality and added value, engage in gastronomes and tasting exhibitions, dissemination and publicity actions, as well as enhance the quality of the products and markets based on the customer needs. The primary focus of this publication is to observe and analyze the data retrieved from various tasting exhibitions of the AliAmbra project, with a target goal of improving customer experience and product quality.

Artificial intelligence methods and application have recently shown great contribution in modeling and prediction of the hydrological processes, climate change, and earth systems. Among them, deep learning and machine learning methods mainly have reported being essential for achieving higher accuracy, robustness, efficiency, computation cost, and overall model performance. This paper presents the state of the art of machine learning and deep learning methods and applications in this realm and the current state, and future trends are discussed. The survey of the advances in machine learning and deep learning are presented through a novel classification of methods. The paper concludes that deep learning is still in the first stages of development, and the research is still progressing. On the other hand, machine learning methods are already established in the fields, and novel methods with higher performance are emerging through ensemble techniques and hybridization.

Measurements of systems taken along a continuous functional dimension, such as time or space, are ubiquitous in many fields, from the physical and biological sciences to economics and engineering. Such measurements can be viewed as realisations of an underlying smooth process sampled over the continuum. However, traditional methods for independence testing and causal learning are not directly applicable to such data, as they do not take into account the dependence along the functional dimension. By using specifically designed kernels, we introduce statistical tests for bivariate, joint, and conditional independence for functional variables. Our method not only extends the applicability to functional data of the Hilbert-Schmidt independence criterion (HSIC) and its d-variate version (d-HSIC), but also allows us to introduce a test for conditional independence by defining a novel statistic for the conditional permutation test (CPT) based on the Hilbert-Schmidt conditional independence criterion (HSCIC), with optimised regularisation strength estimated through an evaluation rejection rate. Our empirical results of the size and power of these tests on synthetic functional data show good performance, and we then exemplify their application to several constraint- and regression-based causal structure learning problems, including both synthetic examples and real socio-economic data.

Pipeline integrity is an important area of concern for the oil and gas, refining, chemical, hydrogen, carbon sequestration, and electric-power industries, due to the safety risks associated with pipeline failures. Regular monitoring, inspection, and maintenance of these facilities is therefore required for safe operation. Large standoff magnetometry (LSM) is a non-intrusive, passive magnetometer-based mea- surement technology that has shown promise in detecting defects (anomalies) in regions of elevated mechanical stresses. However, analyzing the noisy multi-sensor LSM data to clearly identify regions of anomalies is a significant challenge. This is mainly due to the high frequency of the data collection, mis-alignment between consecutive inspections and sensors, as well as the number of sensor measurements recorded. In this paper we present LSM defect identification approach based on ma- chine learning (ML). We show that this ML approach is able to successfully detect anomalous readings using a series of methods with increasing model complexity and capacity. The methods start from unsupervised learning with "point" methods and eventually increase complexity to supervised learning with sequence methods and multi-output predictions. We observe data leakage issues for some methods with randomized train/test splitting and resolve them by specific non-randomized splitting of training and validation data. We also achieve a 200x acceleration of support-vector classifier (SVC) method by porting computations from CPU to GPU leveraging the cuML RAPIDS AI library. For sequence methods, we develop a customized Convolutional Neural Network (CNN) architecture based on 1D convolutional filters to identify and characterize multiple properties of these defects. In the end, we report the scalability of the best-performing methods and compare them, for viability in field trials.

Multimodal remote sensing data classification can enhance the model’s ability to distinguish land features through multimodal data fusion. In this context, how to help models understand the relationship between multimodal data and target tasks is the focus of researchers. Inspired by human feedback learning mechanisms, causal reasoning mechanisms, and knowledge induction mechanisms, this paper integrates causal learning, reinforcement learning, and meta learning into a unified remote sensing data classification framework and proposed the causal meta-reinforcement learning (CMRL). First, based on feedback learning mechanisms, we have overcome the limitations of traditional implicit optimization of fusion features and customized a reinforcement learning environment for multimodal remote sensing data classification tasks. Through feedback interactive learning between agents and the environment, we help them understand the complex relationships between multimodal data and labels, thereby achieving full mining of multimodal complementary information. Second, based on the causal inference mechanism we designed causal distribution prediction action, classification reward, and causal intervention reward, capturing pure causal factors in multimodal data and cutting off false statistical associations between non-causal factors and class labels. Finally, based on the knowledge induction mechanism, we designed a bi-layer optimization mechanism based on meta-learning. By constructing a meta training task and meta validation task simulation model in the generalization scenario of unseen data, we helped the model induce cross-task shared knowledge, thereby improving its generalization ability for unseen multimodal data. The experimental results on multiple sets of multimodal datasets show that the proposed method achieved state-of-the-art performance in multimodal remote sensing data classification tasks.

Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify the physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the proposed models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from 8 volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides the accuracy performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. Our results indicate that the proposed method will provide the medical doctors and trainers a promising way to precisely track and understand a patient’s physical activities for better treatment.

In training a deep learning system to perform audio transcription, two practical problems may arise. Firstly, most datasets are weakly labelled, having only a list of events present in each recording without any temporal information for training. Secondly, deep neural networks need a very large amount of labelled training data to achieve good quality performance, yet in practice it is difficult to collect enough samples for most classes of interest. In this paper, we propose factorising the final task of audio transcription into multiple intermediate tasks in order to improve the training performance when dealing with this kind of low-resource datasets. We evaluate three data-efficient approaches of training a stacked convolutional and recurrent neural network for the intermediate tasks. Our results show that different methods of training have different advantages and disadvantages.

Within rehabilitation, there is great need for a simple method to monitor wheelchair use, especially whether it is active or passive. For this purpose, an existing measurement technique was extended with a method for detecting self or attendant pushed wheelchair propulsion. The validity was determined by comparison with manual annotation of wheelchair use. Twenty-four amputation and stroke patients completed a semi-structured course of active and passive wheelchair use. Based on a machine learning approach, a method was developed that detected the type of movement. The machine learning method was trained based on the data of a single wheel sensor as well as a setup using an additional sensor on the frame. The method showed high accuracy (F1 = 0.886, frame and wheel sensor) even if only a single wheel sensor was used (F1 = 0.827). The developed and validated measurement method is ideally suited to easily determine wheelchair use and the corresponding activity level of patients in rehabilitation.

Road collisions are among the world’s critical issues causing many casualties, deaths, and economic losses. To deal with this scenario, precise analysis is required. In this paper, we use Machine Learning (ML) techniques to assess and predict road traffic collisions. An ensemble machine learning model with two layers was developed to analyze simulation data from the driving simulator. The first (base) layer integrates supervised learning techniques namely, k- Nearest Neighbors (k-NN) AdaBoost, Naive Bayes (NB), and Decision Trees (DT). The second layer predicts road collisions by combining the base layer outputs by employing the stacking ensemble method using logistic regression as a meta-classifier. The data for this study were imbalanced; therefore, we performed a SMOTE resampling strategy to handle the dataset imbalance. To simplify the model, the PSO algorithm was used to select the most important features in our dataset. The two-layer ensemble model proposed had the best outcomes with an accuracy of 88%, F1 score of 83%, and an AUC of 86% as compared with k-NN, DT, NB, and AdaBoost.

Machine learning plays a key role in present day crime detection, analysis and prediction. The goal of this work is to propose methods for predicting crimes classified into different categories of severity. We implemented visualization and analysis of crime data statistics in recent years in the city of Boston. We then carried out a comparative study between two supervised learning algorithms, which are decision tree and random forest based on the accuracy and processing time of the models to make predictions using geographical and temporal information provided by splitting the data into training and test sets. The result shows that random forest as expected gives a better result by 1.54% more accuracy in comparison to decision tree, although this comes at a cost of at least 4.37 times the time consumed in processing. The study opens doors to application of similar supervised methods in crime data analytics and other fields of data science

Soccer is type of sport that carries a high risk of injury. Injury is not only cause in the unlucky soccer carrier and also team performance as well as financial effects can be worse since soccer is a team-based game. The duration of recovery from a soccer injury typically relies on its type and severity. Therefore, we conduct this research in order to predict the probability of players injury type using machine learning technologies in this paper. Furthermore, we compare different machine learning models to find the best fit model. Supervised classification machine learning models are applied in this paper. We gathered information about 54 professional soccer players who are playing in the top five European leagues based on their career history.

The fossil record is notorious for being incomplete and distorted, frequently conditioning the type of knowledge that can be extracted from it. In many cases, this often leads to issues when performing complex statistical analyses, such as classification tasks, predictive modelling, and variance analyses, such as those used in Geometric Morphometrics. Here different Generative Adversarial Network architectures are experimented with, testing the effects of sample size and domain dimensionality on model performance. For model evaluation, robust statistical methods were used. Each of the algorithms were observed to produce realistic data. Generative Adversarial Networks using different loss functions produced multidimensional synthetic data significantly equivalent to the original training data. Conditional Generative Adversarial Networks were not as successful. The methods proposed are likely to reduce the impact of sample size and bias on a number of statistical learning applications. While Generative Adversarial Networks are not the solution to all sample-size related issues, combined with other pre-processing steps these limitations may be overcome. This presents a valuable means of augmenting geometric morphometric datasets for greater predictive visualization.

In response to the growing challenges in drone security and airspace management, this study introduces an advanced drone classifier, capable of detecting and categorizing Unmanned Aerial Vehicles (UAVs) based on acoustic signatures. Utilizing a comprehensive database of drone sounds across EU-defined classes (C0 to C3), this research leverages machine learning (ML) techniques for effective UAV identification. The study primarily focuses on the impact of data augmentation methods—pitch shifting, time delays, harmonic distortion, and ambient noise integration—on classifier performance. These techniques aim to mimic real-world acoustic variations, thus enhancing the classifier's robustness and practical applicability. Results indicate that moderate levels of augmentation significantly improve classification accuracy. However, excessive application of these methods can negatively affect performance. The study concludes that sophisticated acoustic data augmentation can substantially enhance ML-driven drone detection, providing a versatile and efficient tool for managing drone-related security risks. This research contributes to UAV detection technology, presenting a model that not only identifies but also categorizes drones, underscoring its potential for diverse operational environments.

Agricultural Big Data is a set of technologies that allows responding to the challenges of the new data era. In conjunction with machine learning, farmers can use data to address different problems such as farmers' decision-making, crops, weeds, animal research, land, food availability and security, weather, and climate change. The purpose of this paper is to synthesize the evidence regarding the challenges involved in implementing machine learning in Agricultural Big Data. We conducted a Systematic Literature Review applying the PRISMA protocol. This review includes 30 papers, published from 2015 to 2020. We develop a framework that summarizes the main challenges encountered, the use of machine learning techniques, as well as the main technologies used. A major challenge is the design of Agricultural Big Data architectures, due to the need to modify the set of technologies adapting the machine learning techniques, as the volume of data increases.

Physics-informed machine learning (PIML) refers to the emerging area of extracting physically relevant solutions to complex multiscale modeling problems and has enjoyed significant interest from the research community. This paper discusses the recent critical advancements in the PIML domain. Novel methods and applications of domain decomposition in physics-informed neural networks (PINN) in particular are highlighted. Additionally, we explore recent Works toward utilizing neural operator learning to intuit relationships in physics systems traditionally modeled by sets of complex governing equations and solved with expensive differentiation techniques. Finally, expansive applications of traditional physics-informed machine learning and potential limitations are discussed. In addition to summarizing recent work, we propose a novel taxonomic structure to catalog physics-informed machine learning based on how the physics-information is derived and injected into the machine learning process. The taxonomy assumes the explicit objectives of facilitating interdisciplinary collaboration in methodology, thereby promoting a wider characterization of what types of physics-problems are served by the physics-informed learning machines, and assisting in identifying apt targets for future work. To summarize, the major twofold goal of this work is to summarize recent advancements and introduce a taxonomic catalog for applications of physics-informed machine learning.

A data-driven analysis method known as dynamic mode decomposition (DMD) approximates the linear Koopman operator on projected space. In the spirit of Johnson-Lindenstrauss Lemma, we will use random projection to estimate the DMD modes in reduced dimensional space. In practical applications, snapshots are in high dimensional observable space and the DMD operator matrix is massive. Hence, computing DMD with the full spectrum is infeasible, so our main computational goal is estimating the eigenvalue and eigenvectors of the DMD operator in a projected domain. We will generalize the current algorithm to estimate a projected DMD operator. We focus on a powerful and simple random projection algorithm that will reduce the computational and storage cost. While clearly, a random projection simplifies the algorithmic complexity of a detailed optimal projection, as we will show, generally the results can be excellent nonetheless, and quality understood through a well-developed theory of random projections. We will demonstrate that modes can be calculated for a low cost by the projected data with sufficient dimension.

This paper offers a comprehensive examination of the process involved in developing and automating supervised end-to-end machine learning workflows for forecasting and classification purposes. It offers a complete overview of the components (i.e. feature engineering, model selection, etc), principles (i.e. bias-variance decomposition, model complexity, overfitting, model sensitivity to feature assumptions and scaling, output interpretability, etc), models (i.e. neural networks, regression models, etc), methods (i.e. Cross-Validation, data augmentation, etc), metrics (i.e. Mean Squared Error, F1-score, etc) and tools that rule most supervised learning applications with numerical and categorical data, as well as their integration, automation, and deployment. The end goal and contribution of this paper is the education and guidance of the non-AI expert academic community over complete and rigorous machine learning pipelines and data science practices, from problem scoping to design and state-of-the-art automation tools, including basic principles and reasoning in the choice of methods. The paper delves into the critical stages of supervised machine learning workflow development, many of which are often omitted by researchers due to brevity, and covers foundational concepts essential for understanding and optimizing a functional machine learning workflow, thereby offering a holistic view of task-specific application development for applied researchers who are not AI experts.

Real estate market analysis and place-based decision-making can both benefit from understanding house price development. Although considerable amounts of interest have been devoted to housing price modelling, the assessment of house price fluctuation still requires further comparing studies. Housing price prediction is challenging as contributing factors are quite dynamic and subject to a variety of regulating elements. The future understanding of the housing market trends not only provides sufficient customers’ investment trust potential but also enables the financial support to progress more realistic in advance. In this study, a comprehensive data-informed framework is developed to investigate and anticipate real estate house prices using historical data by combining explanatory features. We examined about 500 houses in the Boston area as a case study and discussed how the increase in housing prices could vary by each of the contributing components. Fourteen Machine Learning (ML) regressors imply to the dataset and lead to a comparative study of the accuracy of all the models. ML-based regressors forecast real estate home prices as a function of thirteen influencing factors. The most informative features were also selected by conducting the Permutation Feature Importance technique on all the features The study provides a comprehensive tool for evaluating the robustness and efficiency of ML models for housing price predictions. The results highlighted Random Forest as the best model has an R2 equals to 0.88 and Voting Regressor as the second highest rated model has R2 equals to 0.87. Results of multivariate exploratory data analysis also implied that the average number of rooms and percentage of the lower status of the population have the most significant impact on the price range predictions.

Today, across the most critical problems faced by hospitals and health centers are those caused by the existence of patients who do not attend their appointments. Among others, this practice generates waste of resources and increases the patients’ waiting list. To handle these problems, hospitals are actively trying to implement methods to reduce the idle time caused by patient no-shows. Many scheduling systems developed require predicting whether a patient will show up for an appointment or not. Although, a challenging problem resides in obtaining these estimates precisely. The goal of this work is to analyze how objective factors influence a patient not to attending their appointment, to identify the main causes that contribute to a patient’s decision, and to be able to predict whether or not the patient will attend the scheduled appointment. As a result, the obtained model is tested on a real dataset collected in a health center linked to the University of Vale do Itajaí (UNIVALI), which includes 25 features and about 5000 samples. The algorithm that produced the best results for the available dataset is the Random Forest classifier. It reveals the best recall rate (0.91), since it measures the ability of a classifier to find all the positive instances and achieves a receiver operating characteristic curve rate of 0.969.

This paper presents a new framework to classify floor plan elements and represent them in a vector format. Unlike existing approaches using image-based learning frameworks as the first step to segment the image pixels, we first convert the input floor plan image into vector data and utilize graph neural network. Our framework consists of three steps. (1) image pre-processing and vectorization of the floor plan image. (2) region adjacency graph conversion. (3) graph neural network on converted floor plan graphs. Our approach is able to capture different types of indoor elements including basic elements such as walls, doors, and symbols as well as spatial elements such as rooms and corridors. In addition, the proposed method can also detect element shapes. Experimental results show that our framework can classify indoor elements with an F1 score of 95%, with scale and rotation invariance. Furthermore, we propose a new graph neural network model that takes the distance between nodes into account, which is a valuable feature of spatial network data.

Recently, deep learning-based image quality enhancement models have been proposed to improve the perceptual quality of distorted synthesized views impaired by compression and Depth Image Based Rendering (DIBR) process in multiview video systems. However, due to the lack of multi-view video plus depth data, the training data for quality enhancement models is small, which limits the performance and progress of these models. Augmenting the training data to enhance the Synthesized View Quality Enhancement (SVQE) models is a feasible solution. In this paper, we suggest a deep learning-based SVQE model using more synthetic Synthesized View Images (SVIs). To simulate the irregular geometric displacement of DIBR distortion, a random irregular polygon-based SVI synthesis method is proposed based on existing massive RGB/RGBD data, and a synthetic synthesized view database is constructed, which includes synthetic SVIs and DIBR distortion masks. Moreover, to further guide the SVQE models to focus more precisely on DIBR distortion, the DIBR distortion mask prediction network which could predict the position and variance of DIBR distortion is embedded into the SVQE models. The experimental results demonstrate that by pretraining on the synthetic SVI database, the performance of the existing SVQE models could be greatly promoted. In addition, by introducing the DIBR distortion mask prediction network, the SVI quality could be further enhanced.

Water quality assessments are crucial for human health and environmental safeguards. The utilization of a subset of artificial intelligence such as Machine Learning (ML) presents significant impacts to enhance the prediction and classification of water quality. In this research, a set of diverse ML algorithms was evaluated to handle a comprehensive dataset of water quality measurements over an extended period. The aim was to develop a robust approach for accurately forecasting water quality. This approach employed machine learning classifiers such as Logistic Regression (LR), Support Vector Machine (SVM), Stochastic Gradient Descent (SGD), K-Nearest Neighbors (KNN), Gaussian Process Classification (GPC), Gaussian Naive Bayes (GNB), Random Forest (RF), Decision Tree (DT), XGBoost, and Multilayer Perceptron (MLP). The water quality parameters assessed for pH, hardness, solids, chloramines, sulfate, conductivity, organic carbon, trihalomethanes and turbidity. The XGBoost model exhibited the highest accuracy of 89.47% among the classifiers and Stacked Ensemble Classifiers (SEC) improved the prediction further to 92.98%. The findings suggest that XGBoost and the SEC hold promise as reliable approaches for water quality assessments in contrast of artificial intelligence.

Hepatitis C presents a significant global health challenge, necessitating early diagnosis and precise severity classification for timely medical intervention. This study explores the application of machine learning techniques to predict the severity of hepatitis C, leveraging an extensive dataset. Our approach encompasses rigorous data preprocessing, advanced model development, and fine-tuned hyperparameter optimization to ensure accurate and reliable predictions. We evaluated four classification models: Logistic Regression, Random Forest, Gradient Boosting, and Support Vector Machine (SVM), comparing their accuracy in classifying patients. The results showed that the Random Forest and Gradient Boosting models outperformed the others with an accuracy of approximately 93.50%, demonstrating their potential in assisting Hepatitis C diagnosis. Further model enhancements through hyperparameter tuning and feature engineering can improve the precision of Hepatitis C diagnosis, contributing to better patient care.

Data collection and analysis are becoming more and more important in a variety of application domains as long as the novel technologies advance. At the same time, we are experiencing a growing need for human-machine interaction with expert systems pushing research through new knowledge representation models and interaction paradigms. In particular, in the last years eHealth - that indicates all the health-care practices supported by electronic elaboration and remote communications - calls for the availability of smart environment and big computational resources. The aim of this paper is to introduce the HOLMeS (Health On-Line Medical Suggestions) framework. The introduced system proposes to change the eHealth paradigm where a trained machine learning algorithm, deployed on a cluster-computing environment, provides medical suggestion via both chat-bot and web-app modules. The chat-bot, based on deep learning approaches, is able to overcome the limitation of biased interaction between users and software, exhibiting a human-like behavior. Results demonstrate the effectiveness of the machine learning algorithms showing 74.65% of Area Under ROC Curve (AUC) when first-level features are used to assess the occurrence of different prevention pathways. When disease-specific features are added, HOLMeS shows 86.78% of AUC achieving a more specific prevention pathway evaluation.

Pilots can be one of the factors in many air traffic accidents. When one or both pilots are impaired (e.g. fatigue, drunk), disabled, capable but wrong-headed, don’t have sufficient training, distracted, miscommunicate with the air traffic controller, or follow wrong instructions from the air traffic controller, the risk of accident will increase dramatically. In some of these cases, the risk can be mitigated by using big data and machine learning. The system will collect and analyze large amount of data about the state of the aircraft, e.g., the flight path, the immediate environment around the aircraft, the weather and terrain information, and the pilots’ input to control the aircraft. Additional sensors such as eye tracking devices and biological monitor can also be added to determine the condition of the pilots. If the pilots’ input do not match proper reaction to the situation or the pilots are impaired, the learning machine will first provide an advisory to the pilots. If both pilots are impaired or incapable, a warning will be sent to the flight attendants and air traffic controllers so that they can take appropriate actions. The learning machine will be trained by both accident database and an automatic training system.

The comprehensibility of good predictive models learned from high-dimensional gene expression data is attractive because it can lead to biomarker discovery. Several good classifiers provide comparable predictive performance but differ in their abilities to summarize the observed data. We extend a Bayesian Rule Learning (BRL-GSS) algorithm, previously shown to be a significantly better predictor than other classical approaches in this domain. It searches a space of Bayesian networks using a decision tree representation of its parameters with global constraints, and infers a set of IF-THEN rules. The number of parameters and therefore the number of rules are combinatorial to the number of predictor variables in the model. We relax these global constraints to a more generalizable local structure (BRL-LSS). BRL-LSS entails more parsimonious set of rules because it does not have to generate all combinatorial rules. The search space of local structures is much richer than the space of global structures. We design the BRL-LSS with the same worst-case time-complexity as BRL-GSS while exploring a richer and more complex model space. We measure predictive performance using Area Under the ROC curve (AUC) and Accuracy. We measure model parsimony performance by noting the average number of rules and variables needed to describe the observed data. We evaluate the predictive and parsimony performance of BRL-GSS, BRL-LSS and the state-of-the-art C4.5 decision tree algorithm, across 10-fold cross-validation using ten microarray gene-expression diagnostic datasets. In these experiments, we observe that BRL-LSS is similar to BRL-GSS in terms of predictive performance, while generating a much more parsimonious set of rules to explain the same observed data. BRL-LSS also needs fewer variables than C4.5 to explain the data with similar predictive performance. We also conduct a feasibility study to demonstrate the general applicability of our BRL methods on the newer RNA sequencing gene-expression data.

Machine learning (ML) is increasingly useful as data grows in volume and accessibility as it can perform tasks (e.g. categorisation, decision making, anomaly detection, etc.) through experience and without explicit instruction, even when the data are too vast, complex, highly variable, full of errors to be analysed in other ways , . Thus, ML is great for natural language, images, or other complex and messy data available in large and growing volumes. Selecting a ML algorithm depends on many factors as algorithms vary in supervision needed, tolerable error levels, and ability to account for order or temporal context, among many other things. Importantly, ML methods for explicitly ordered or time-dependent data struggle with errors or data asymmetry. Most data are at least implicitly ordered, potentially allowing a hidden `arrow of time’ to affect non-temporal ML performance. This research explores the interaction of ML and implicit order by training two ML algorithms on Twitter data before performing automatic classification tasks under conditions that balance volume and complexity of data. Results show that performance was affected, suggesting that researchers should carefully consider time when selecting appropriate ML algorithms, even when time is only implicitly included.

Industrial pig farming is associated with negative technological pressure on the bodies of pigs. Leg weakness and lameness are the sources of significant economic loss in raising pigs. Therefore, it is important to identify predictors of limb condition. This work presents assessments of the state of limbs using indicators of growth and meat characteristics of pigs based on machine learning algorithms. We have evaluated and compared the accuracy of prediction for several ML classification algorithms (Random Forest, K-Nearest Neighbors, Artificial Neural Networks, C50Tree, Support Vector Machines, Naive Bayes, Generalized Linear Models, Boost, and Linear Discriminant Analysis) and have identified the Random Forest and K-Nearest Neighbors as the best performing algorithms for predicting pig leg weakness using a small set of simple measurements that can be taken at an early stage of animal development. Muscle Thickness, Back Fat amount, and Average Daily Gain serve as significant predictors of conformation of pig limbs. Our work demonstrates the utility and relative ease of using machine learning algorithms to assess the state of limbs in pigs based on growth rate and meat characteristics.

Chronic venous disease (CVD) occurs in a substantial proportion of the world's population. If the onset of CVD looks like a cosmetic defect, then over time, it can develop into serious problems that require surgical intervention. The aim of the work is to use deep learning (DL) methods for automatic classification of the stage of CVD for self-diagnosis of a patient by using the image of the patient’s legs. The required for DL algorithms images of legs with CVD were obtained by using Internet Data Mining. For images preprocessing, the binary classification problem “legs - no legs” was solved based on Resnet50 with accuracy 0.998. The application of this filter made it possible to collect a data set of 11,118 good quality leg images with various stages of CVD. For classification of various stages of CVD according to CEAP classification, the multi classification problem was set and resolved by using two neural networks with completely different architecture - Resnet50 and DeiT. The model based on DeiT without any tuning shows better results than the model based on Resnet50 (precision = 0.770 (DeiT) and 0.615 (Resnet50)). To demonstrate the results of the work, a telegram bot was developed, in which fully functioning DL algorithms are implemented. This bot allows evaluating the condition of the patient's legs with a fairly good accuracy for the CVD classification.

Power disaggregation aims at determining the appliance-by-appliance electricity consumption leveraging upon a single meter only, which measures the entire power demand. Data-driven procedures based on Factorial Hidden Markov Models have been proven remarkable results on energy disaggregation. Nevertheless, those procedures have various weaknesses: there is a scalability problem as the number of devices to observe raises and the algorithmic complexity of the inference step is severe. DNN architectures, such as Convolutional Neural Networks, have demonstrated to be a viable solution to deal with FHMMs shortcomings. Nonetheless, there are two significant limitations: a complicated and time-consuming training system based on back-propagation has to be employed to estimates the neural architecture parameters, and large amounts of training data covering as many operation conditions as possible need to be collected to attain top performances. In this work, we aim to overcome those limitations by leveraging upon the unique and useful characteristics of the extreme learning machine technique, which is based on a collection of randomly chosen hidden units and analytically defined output weights. Experiment evaluation has been conducted using the UK-DALE corpus. We find that the suggested approach achieves similar performances to recently proposed ANN-based methods and outperforms FHMMs. Besides, our solution generalises well to unseen houses.

The necessity for effective intrusion detection systems (IDS) in big data environments has grown critical due to the rising prevalence of big data systems and the rising amount of security threats. Applying conventional intrusion detection methods to the enormous and intricate data produced by big data ecosystems presents difficulties. Deep learning, has proven to be exceptionally adept at deciphering complex, large-scale data. This study proposes a Deep Learning-based Intrusion Detection System (IDS) created specifically for the Big Data environment. Traditional intrusion detection systems struggle to efficiently identify and prevent cyber threats due to the ever-increasing volume and complexity of data in current networks. We suggest an IDS that makes use of Deep Learning (CNN, LSTM, GAN, etc.), to address this problem. These cutting-edge neural network topologies give the system the ability to process and analyze massive amounts of data for precise intrusion detection. Data collection, data preprocessing, feature engineering, DL model training, intrusion detection, warning generation, and reaction are some of the crucial parts of the suggested IDS. The data collection module mines the Big Data environment for network traffic, system logs, and security event information. To make sure the data is suitable for analysis, it is preprocessed. To improve the detection abilities of the DL models, important features are extracted from the data using feature engineering approaches.

Precision medicine emphasizes fine-grained diagnostics, taking individual variability into account to enhance treatment effectiveness. Parkinson's Disease (PD) heterogeneity among individuals is a proof that disease subtypes exist, and assigning individuals to subgroups is necessary for a better understanding of disease mechanisms and designing precise treatment approaches. The purpose of this study was to identify PD subtypes using RNA-Seq data in a combined pipeline including unsupervised machine learning, bioinformatics, and network analysis. 210 post mortem brain RNA-Seq samples from PD (n = 115) and Normal Controls (NC, n = 95) were obtained with a systematic data retrieval following PRISMA statements and a fully data-driven clustering pipeline was performed to identify PD subtypes. Bioinformatics and Network analyses were performed to characterize the disease mechanisms of the identified PD subtypes and to identify target genes for drug repurposing. Two PD clusters were identified and 42 DEGs were found (p.adjusted ≤ 0.01). PD clusters had significantly different gene network structures (p < 0.0001) and phenotype-specific disease mechanisms, highlighting the differential involvement of the Wnt/β-catenin pathway regulating adult neurogenesis. NEUROD1 was identified as a key regulator of gene networks and ISX9 and PD98059 were identified as NEUROD1-interacting compounds with disease-modifying potential, reducing the effects of dopaminergic neurodegeneration. This hybrid data analysis approach could enable precision medicine applications by providing insights for the identification and characterization of pathological subtypes. This workflow has proven useful on PD brain RNA-Seq, but its application to other neurodegenerative diseases is encouraged.

Machine learning has become an increasingly ubiquitous technology, as big data continues to inform and influence everyday life and decision-making. Currently in healthcare, as well as in most other industries, the two most prevalent machine learning paradigms are supervised learning and transfer learning. Both practices rely on large-scale, manually annotated datasets to train increasingly complex models. However, the requirement of data to be manually labeled leaves an excess of unused, unlabeled data available in both public and private data repositories. Self-supervised learning (SSL) is a growing area of machine learning that has the ability to take advantage of unlabeled data. Contrary to other machine learning paradigms, SSL algorithms create artificial supervisory signals from unlabeled data and pretrain algorithms on these signals. The aim of this review is two-fold: firstly, we provide a formal definition of SSL, divide SSL algorithms into their four unique subsets, and review the state-of-the-art published in each of those subsets between the years of 2014-2020. Second, this work surveys recent SSL algorithms published in healthcare, in order to provide medical experts with a clearer picture of how they can integrate SSL into their research, with the objective of leveraging unlabeled data.

Delayed cancer detection is one of the common causes of poor prognosis in case of many cancers including the cancers of the oral cavity. Despite improvement and development of new and efficient gene therapy treatments, very little has been done to algorithmically assess the impedance of these carcinomas. In this work, we attempt to annotate viable attributes in oral cancer gene datasets for identification of gingivobuccal cancer (GBC). We further apply supervised and unsupervised machine learning methods to the gene datasets revealing key candidate attributes for GBC prognosis. Our work highlights the importance of automated identification of key genes responsible for GBC that could perhaps be easily replicated to other forms of oral cancer detection.

The modeling and prediction of crop seed yield can be a vital improvement in the precision agriculture industry as it provides reliable assessments of the effectiveness of agro-traits. Here, multiple machine learning (ML) techniques are established for predicting sesame (Sesamum indicum L.) seed yield (SSY) and incorporating agro-morphological features. Models utilized for coupled PCA-ML (Principal component analysis-Machine Learning) methods were compared with original ML models to evaluate predicted efficiency. The Gaussian process regression (GPR) and Radial basis function neural network (RBF-NN) models exhibited the most accurate SSY predictions with determination coefficients or R2 values of 0.99 and 0.91, respectively. The root-mean-square error (RMSE) for the ML models fluctuated between 0 to 0.30 t/ha (metric tons/hectare) for the varied modeling process phases. Estimation of sesame seed yield with coupled PCA-ML models improved performance accuracy. The K-fold process suggested the utilization of datasets with the lowest error rates to ensure the continued accuracy of GPR and RBF models. Sensitivity analysis revealed the capsule number per plant (CPP), seed number per capsule (SPC), and 1000-seed weight (TSW) were the most significant seed yield determinants.

Boolean functions and networks are commonly used in the modeling and analysis of complex biological systems, and this paradigm is highly relevant in other important areas in data science and decision making, such as in the medical field and in the finance industry. In a Boolean model, the truth state of a variable is either 0 or 1 at a given time. Despite its apparent simplicity, Boolean networks are surprisingly relevant in many areas of application such as in bioinformatics to model gene expressions and interactions. In the latter case, a gene is either ``on" or ``off" depending on its expression level. Despite the promising utility of Boolean modeling, in most practical applications the Boolean network is not known. Automated learning of a Boolean network and Boolean functions, from data, is a challenging task due in part to the large number of unknowns (including both the structure of the network and the functions) to be estimated, for which a brute force approach would be exponentially complex. In this paper we develop a new information theoretic methodology that we show to be significantly more efficient than previous approaches. Building on the recently developed optimal causation entropy principle (oCSE), that we proved can correctly infer networks distinguishing between direct versus indirect connections, we develop here an efficient algorithm that furthermore infers a Boolean network (including both its structure and function) based on data observed from the evolving states at nodes. We call this new inference method, Boolean optimal causation entropy (BoCSE), which we will show that our method is both computationally efficient and also resilient to noise. Furthermore, it allows for selection of a set of features that best explains the process, a statement that can be described as a networked Boolean function reduced order model. We highlight our method to the feature selection in several real-world examples: (1) diagnosis of urinary diseases, (2) Cardiac SPECT diagnosis, (3) informative positions in the game Tic-Tac-Toe, and (4) risk causality analysis of loans in default status. Our proposed method is effective and efficient in all examples.

In the context of developing machine learning models, until and unless we have the required data engineering and machine learning development competencies as well as the time to train and test different machine learning models and tune their hyperparameters, it is worth trying out the automatic machine learning features provided by several cloud-based and cloud-agnostic platforms. This paper explores the possibility of generating automatic machine learning models with low-code experience. We have developed criteria to compare different machine learning platforms for generating automatic machine learning models and presenting their results. Thereafter, lessons learned by developing automatic machine learning models from a sample dataset across four different machine learning platforms were elucidated. We have also interviewed machine learning experts to conceptualize their domain-specific problems that automatic machine learning platforms could address. Results showed that automatic machine learning platforms could provide a fast track for organizations seeking digitalization of their businesses. Automatic machine learning platforms help produce results, especially for time-constrained projects where resources are lacking. The contribution of this paper is in the form of a lab experiment in which we demonstrate how low-code platforms could provide a viable option to many business cases and henceforth provide a lane that is faster than the usual hiring and training of already scarce data scientists and to analytics projects that suffer from overruns.

In a computing context, cybersecurity is undergoing massive shifts in technology and its operations in recent days, and data science is driving the change. Extracting security incident patterns or insights from cybersecurity data and building corresponding data-driven model, is the key to make a security system automated and intelligent. To understand and analyze the actual phenomena with data, various scientific methods, machine learning techniques, processes, and systems are used, which is commonly known as data science. In this paper, we focus and briefly discuss cybersecurity data science, where the data is being gathered from relevant cybersecurity sources, and the analytics complement the latest data-driven patterns for providing more effective security solutions. The concept of cybersecurity data science allows making the computing process more actionable and intelligent as compared to traditional ones in the domain of cybersecurity. We then discuss and summarize a number of associated research issues and future directions. Furthermore, we provide a machine learning-based multi-layered framework for the purpose of cybersecurity modeling. Overall, our goal is not only to discuss cybersecurity data science and relevant methods but also to focus the applicability towards data-driven intelligent decision making for protecting the systems from cyber-attacks.

Background: Ongoing molecular profiling studies enabled by advances in biomedical technologies are producing vast amounts of ‘omic’ data for early detection, monitoring, and prognosis of diverse diseases. A major common limitation is the scarcity of biological samples, necessitating integrative modeling frameworks that can make optimal use of available data for disease classification tasks. Related data sets are often available from different studies, but may have been generated using different technology platforms. Thus, there is a critical need for flexible modeling methods that can handle data from diverse sources to facilitate the discovery of robust biomarkers that underlie disease regulatory processes. Results: In this paper, we introduce a novel framework called Knowledge Augmented Rule Learning (KARL), which incorporates two sources of knowledge, domain, and data, for pattern discovery from small and high-dimensional datasets, such as transcriptomic data. We propose KARL as a transfer rule learning framework in which knowledge of the domain is transferred to the learning process on data in order to 1) improve the reliability of the discovered patterns, and 2) study the knowledge of the domain when used along with data for modeling. In this work, we generated KARL models on gene expression datasets for five types of cancer, including brain, breast, colon, lung, and prostate. As our knowledge of the domain, we used the Ingenuity Knowledge Base (IKB) to extract genes related to hallmarks of cancer and annotated these prior relationships before learning classifiers from these datasets. Conclusions: Our results show that KARL produces, on average, rule models that are more robust classifiers than the baseline without such background knowledge, for our tasks of cancer prediction using 25 publicly available gene expression datasets. Moreover, KARL helped us learn insights about previously known relationships in these gene expression datasets, along with new relationships not input as known, to enable informed biomarker discovery for cancer prediction tasks. KARL can be applied to modeling similar data from any other domain and classification task. Future work would involve extensions to KARL to handle hierarchical knowledge to derive more general hypotheses to drive biomedicine.

The need for Artificial Intelligence (AI) and Machine Learning (ML) technologies is increasingly being leveraged for optimizing Data centers’ (DCs’) operations as the volume of operations management data increase tremendously. These strategies can assist operators in better understanding their DC operations and making informed decisions up front to preserve service reliability and availability. Aiming at creating models that optimize energy efficiency, identify inefficient resource utilization and scheduling policies, and predict outages. Apart from model hyperparameter tuning, feature selection is a crucial step to identify relevant features with the objective of providing insight into the data, improving performance, and reducing computational expenses. Although several feature selection methods have been discussed in various domains, none have been discussed in the context of the data center. This paper introduces SHapley Additive exPlanation (SHAP), a class of additive feature attribution values-based feature selection that is rarely discussed in literature. We compared the effectiveness of SHAP method with several widely used methods. We used a real DC dataset obtained from the ENEA CRESCO6 cluster with 2,0832 cores to evaluate the methods. To demonstrate the comparison of the methods, we picked the top 10 most important features from each method, the predictions were retrained, and their performance was evaluated using MAE, RMSE, and MPAE. The results show that the SHAP-assisted feature selection performed best and align with human intuition.

No-show appointments in healthcare is a problem faced by medical centers around the world, and understand the factors associated with the no-show behavior is essential. In the last decades, artificial intelligence took place in the medical field and machine learning algorithms can work as a efficient tool to understand the patients behavior and to achieve better medical appointment allocation in scheduling systems. In this work, we provide a systematic literature review (SLR) of machine learning techniques applied to no-show appointments aiming at establishing the current state-of-the-art. Based on a SLR following the Kitchenham methodology, 24 articles were found and analyzed, in which the characteristics of the database, algorithms and performance metrics of each studies were synthesized. Results regarding which factors have a higher impact on missed appointment rates were analyzed too. The results indicate that the most appropriate algorithms for building the models are decision tree algorithms. Furthermore, the most significant determinants of no-show were related to the patients age, whether the patient missed a previous appointment, and the distance between the appointment and the patients scheduling.

The purpose of this study was to detect mind-wandering experienced by pre-service teachers while learning video lecture on physics. The lecture was videotaped and consisted of a live lecture in a classroom. The lecture was about Gauss's law on physics. We investigated whether oculomotor data and eye movements could be used as a marker to indicate the learner’s mind-wandering. Each data was collected in a study in which 24 pre-service teachers (16 females and 8 males) reported self-caught mind-wandering while learning physics video lecture during30 minutes. A Tobii Pro Spectrum (sampling rate: 300Hz) was used to capture their eye-gaze during learning Gauss's law course video. After watching video lecture, we interviewed pre-service teachers about their mind-wandering experience. We first used the self-caught method to capture the mind-wandering timing of pre-service teachers while learning from video lectures. We detected more accurate mind-wandering segments by comparing fixation duration and saccade count. We investigated two types of oculomotor data (blink count, pupil size) and nine eye movements (average peak velocity of saccades; maximum peak velocity of saccades; standard deviation of peak velocity of saccades; average amplitude of saccades; maximum amplitude of saccades; total amplitude of saccades; saccade count/s; fixation duration; fixation dispersion). The result was that the blink count could not be used as a marker for mind-wandering during learning video lectures among them (oculomotor data and eye movements), unlike previous literatures. Based on the results of this study, we identified elements that can be used as mind-wandering markers while learning from video lectures that are similar to real classes, among the oculomotor data and eye movement mentioned in previous literatures. Also, we found that most participants focused on past thoughts and felt unpleasant after experiencing mind-wandering through interview analysis.

Automated spectrum analysis serves as a troubleshooting tool that helps to diagnose faults in wireless networks like difficult signal propagation conditions and coexisting wireless networks. It provides a higher monitoring coverage while requiring less expertise compared to manual spectrum analysis. In this paper, we introduce a data set that can be used to train and evaluate deep learning models, capable of detecting frames from different wireless standards as well as interference between single frames. Since manually labelling a high variety of frames in different environments is too challenging, an artificial data generation pipeline has been developed. The data set consists of 20 000 augmented signal segments, each containing a random number of different Wi-Fi and Bluetooth frames, their spectral image representations and labels that describe the position and type of frame within the spectrogram. The data set contains results of intermediate processing steps that enables the research or teaching community to create new data sets for specific requirements or to provide new interesting examination examples.

Energy efficiency of Data Center (DC) operations heavily relies on IT and cooling systems performance. A reliable and efficient cooling system is necessary to produce a persistent flow of cold air to cool servers that are subjected to constantly increasing computational load due to the advent of IoT- enabled smart systems. Consequently, increased demand for computing power will bring about increased waste heat dissipation in data centers. In order to bring about a DC energy efficiency, it is imperative to explore the thermal characteristics analysis of an IT room (due to waste heat). This work encompasses the employment of an unsupervised machine learning modelling technique for uncovering weaknesses of the DC cooling system based on real DC monitoring thermal data. The findings of the analysis result in the identification of areas for energy efficiency improvement that will feed into DC recommendations. The methodology employed for this research includes statistical analysis of IT room thermal characteristics, and the identification of individual servers that frequently occur in the hotspot zones. A critical analysis has been conducted on available big dataset of ambient air temperature in the hot aisle of ENEA Portici CRESCO6 computing cluster. Clustering techniques have been used for hotspots localization as well as categorization of nodes based on surrounding air temperature ranges. The principles and approaches covered in this work are replicable for energy efficiency evaluation of any DC and thus, foster transferability. This work showcases applicability of best practices and guidelines in the context of a real commercial DC that transcends the set of existing metrics for DC energy efficiency assessment.

In cotton cultivars, an insect that causes irreversible damage is the Spodoptera frugiperda, known as the fall armyworm. Since the visual detection of plants is a burdensome task for human inspection, the spectral information related to plant damage, registered on a spectral scale, can be useful. These measurements, associated with machine learning techniques, produce useful information for a rapid and non-invasive inspection method development. To contribute to this gap fulfillment, this paper proposes a machine learning framework to model the spectral response of cotton plants under the attack of the fall armyworm. Additionally, a theoretical model is presented, built from the results of the machine learning analysis, to infer this damage with up-to-date orbital sensors. The data was composed of the reflectance measurements collected at a cotton field with control plants and plants submitted to Spodoptera frugiperda damage. Their spectral response was recorded with a hand-held spectroradiometer ranging from 350 to 2,500 nm, for eight consecutive days. Different machine learning models were evaluated and the overall best model was defined by accuracies comparisons on a testing-set. A ranking approach was adopted based on the model accuracy, returning the most contributive wavelengths for the classification. Sequentially, an unsupervised neural network (Self-Organizing Map - SOM) was implemented to reduce data-dimensionality and assist in the definition of important spectral regions. The regions were associated with the spectral bands of the two sensors (OLI and MSI) and a theoretical model using a band simulation process with the overall best machine learning model was proposed. The results indicated that the Random Forest (RF) algorithm is the most suitable to predict cotton-plants damaged by insects and that the last day of analysis (8th day) was better to separate it, with F-measure equals 0.912. The ranking approach combined with the SOM method indicated the spectral regions at the red to near-infrared (650 to 1,350 nm) and shortwave infrared (1,570 to 1,640 nm) as the most important regions to the analysis. The proposed theoretical model simulated with the OLI and MSI sensor-bands returned an F-Measure of 0.865 and 0.886, respectively. In conclusion, this framework can be used to map cotton-plants under insect-attack. The theoretical model presents high accuracy to infer the insect-damaged on cotton plants based on multispectral bands from other sensors, being a useful tool for future research that intends to evaluate it in other areas and at different field scales.

Cancer diagnosis stands at the cusp of a profound transformation, driven by the burgeoning capabilities of Artificial Intelligence (AI) and Machine Learning (ML). This comprehensive review illuminates the extraordinary impact of AI and ML in the field, unraveling their multifaceted roles in the realm of oncology. In the arena of cancer diagnosis, AI and ML serve as invaluable allies, empowering healthcare professionals with unparalleled tools for precision and efficiency. Notably, AI's prowess in analyzing medical images, including radiological scans and pathology slides, elevates the early detection of malignancies to new heights. Coupled with its ability to dissect genomic data, AI tailors therapeutic strategies to the individual, promising optimized treatment outcomes. However, the incorporation of AI and ML into clinical practice necessitates a careful navigation of ethical considerations, data privacy, and regulatory landscapes. Safeguarding patient data, ensuring transparency, and addressing algorithmic biases emerge as pivotal challenges that require vigilant attention. Yet, the future of AI and ML in cancer diagnosis is brimming with promise. The integration of multi-modal data, real-time monitoring, and Explainable AI (XAI) methods holds the potential to enrich diagnostic capabilities and engender patient trust. Global collaboration and data sharing initiatives are fostering the development of robust AI models. Furthermore, AI's role in cancer prevention, identifying high-risk individuals and enabling targeted preventive strategies, is poised to revolutionize healthcare. AI and ML are forging a path toward an era of unparalleled accuracy, efficiency, and personalization in cancer diagnosis. Challenges notwithstanding, these technologies bear the promise of fundamentally reshaping patient care, elevating outcomes, and advancing the fight against cancer. The ongoing exploration and responsible implementation of AI and ML in oncology will be pivotal in harnessing their full potential and paving the way for a brighter future for cancer patients.

This paper presents an effective heart disease prediction model through detecting the anomalies, also known as outliers, in healthcare data using the unsupervised K-means clustering algorithm. Most existing approaches for detecting anomalies are based on constructing profiles of normal instances. However, such techniques require an adequate number of normal profiles to justify those models. Our proposed model first evaluates an \textit{optimal} value of K using Silhouette method. Next, it intends to locate anomalies that are far from a certain threshold distance with respect to their clusters. Finally, the five most popular classification techniques such as K-Nearest Neighbor (KNN), Random Forest (RF), Support Vector Machines (SVM), Naive Bayes (NB), and Logistic Regression (LR) are applied to build the resultant prediction model. The effectiveness of the proposed methodology is justified using a benchmark dataset of heart disease.

The main purpose of this paper is to propose a novel optimization model with a new machine learning approach in the first section to achieve the best results in financial institutions in the second section. Since the constancy of efficacy derived from parametric and non-parametric is not significant, this paper provides a scientific assessment at the optimization section and proposes a novel combined parametric and non-parametric model which will be a new experiment in literature perception. A scientific assessment of banks based on a combination of the efficiency measurement method of CCR(Charnes, Cooper and Rhodes model) or CRS(Constant Return to Scale) BCC(Banker, Charnes, and Cooper model) or VRS (Variable Return to Scale) in Data Envelopment Analysis (DEA), as well as Stochastic Frontier Approach (SFA) for 65 banks during Feb to July 2020, are introduced. For analyzing the performance of the parametric and non-parametric approaches we have considered the linear regression and Unreplicated Linear Functional Relationship (ULFR). At the machine learning section, a novel four-layers data mining filtering pre-processes for selected supervised classification as well as unsupervised clustering algorithms to increase the accuracy and to remove unrelated attributes and data are applied. For the four kinds of preprocessing approaches of unsupervised attributes, supervised attributes, supervised instances, and unsupervised instances, we have chosen discretization, attribute selection, stratified remove folds, and resample filters respectively. Based on the nature of the suggested financial institution's dataset and attributes, the most appropriate preprocessing filter in each layer to achieve the highest performance is suggested. Finally, the superior bank, best performance model, and the most accurate algorithm are introduced. The results indicate that the bank number 56 is the superior bank. Among the proposed techniques, the novel recommended CVS compared with CCR-BCC and SFA models, has a more positive correlation with profit risk, and show a higher coefficient of determination values. Sequential Minimal Optimization(SMO) algorithm receives the highest accuracy in all four suggested filtering layers.

Accurate estimation of body fat percentage is essential for various health and fitness applications. Traditional methods for measuring body fat, such as underwater weighing, can be costly and inconvenient. In this study, we apply machine learning techniques to predict body fat percentage using easily accessible body measurements and data. The results highlight the effectiveness of regression models in estimating body fat, with Linear Regression, Ridge Regression, and Bayesian Ridge models achieving R-squared scores of approximately 74%, 73%, and 73%, respectively. These models provide a practical and cost-effective solution for individuals and professionals seeking reliable body fat estimates.

Deep learning dominates automated animal activity recognition (AAR) tasks due to the high performance on large-scale datasets. However, constructing centralised data across diverse farms raises data privacy issues. Federated learning (FL) provides a distributed learning solution to train a shared model by coordinating multiple farms (clients) without sharing their private data. Whereas directly applying FL to AAR tasks often faces two challenges: client-drift during local training and gradient conflicts during global aggregation. In this study, we develop a novel FL framework called FedAAR to achieve AAR with sensor data. Specifically, we devise a prototype-guided local update module to alleviate the client-drift issue, which introduces global prototypes as shared knowledge to force clients to learn consistent features. To reduce gradient conflicts between clients, we design a gradient refinement-based aggregation module to eliminate conflicting components between client gradients during global aggregation, thereby improving the agreement between clients. Experiments are conducted on a public dataset to verify FedAAR’s effectiveness, which consists of 87,621 2-s motion data. The results demonstrate that FedAAR outperforms state-of-the-arts, with precision (75.23%), recall (75.17%), F1-score (74.70%), and accuracy (88.88%), respectively. The ablation experiments also show FedAAR’s robustness against various factors (i.e., different data sizes, communication frequency, and client numbers).

Early diagnosis, prioritization, screening, clustering and tracking of COVID-19 patients, and production of drugs and vaccines are some of the applications that have made it necessary to use a new style of technology to involve, to manage and deal with this epidemic. Strategies backed by artificial intelligence (AI) and the Internet of Things (IoT) have been undeniable to understand how the virus works and try to prevent it from spreading. Accordingly, the main aim of this survey article is to highlight the methods of ML, IoT and the integration of IoT and ML-based techniques in the applications related to COVID-19 from the diagnosis of the disease to the prediction of its outbreak. According to the main findings, IoT provided a prompt and efficient approach of following the disease spread. Most of the studies developed by ML-based techniques for handling COVID-19 based dataset provided performance criteria. The most popular performance criteria, is related to accuracy factor. It can be employed for comparing the ML-based methods with different datasets. According to the results, CNN with SVM classifier, Genetic CNN and pre-trained CNN followed by ResNet, provided highest accuracy values. On the other hand, the lowest accuracy was related to single CNN followed by XGboost and KNN methods.

Despite the importance of few-shot learning, the lack of labeled training data in the real world, makes it extremely challenging for existing machine learning methods as this limited data set does not represent the data variance well. In this research, we suggest employing a generative approach using variational autoencoders (VAEs), which can be used specifically to optimize few-shot learning tasks by generating new samples with more intra-class variations. The purpose of our research is to increase the size of the training data set using various methods to improve the accuracy and robustness of the few-shot face recognition. Specifically, we employ the VAE generator to increase the size of the training data set, including the basic and the novel sets while utilizing transfer learning as the backend. Based on extensive experimental research, we analyze various data augmentation methods to observe how each method affects the accuracy of face recognition. We conclude that the face generation method we proposed can effectively improve the recognition accuracy rate to 96.47% using both the base and the novel sets.

Process safety plays a vital role in the modern process industry. To prevent undesired accidents caused by malfunctions or other disturbances in complex industrial processes, considerable attention has been paid to data-driven fault detection techniques. To explore the underlying manifold structure, manifold learning methods including Laplacian eigenmaps, locally linear embedding, and Hessian eigenmaps have been utilized in data-driven fault detection. However, only the partial local structure information is extracted from the aforementioned methods. In this paper, these typical manifold learning methods are synthesized to find the underlying manifold structure from different angles. A more comprehensive local structure is discovered under a unified framework by constructing an objection optimization function for dimension reduction of the process data. The proposed method takes advantage of different manifold learning methods. Based on the proposed dimension reduction method, a new data-driven fault detection method is developed. Hotelling’s T2 and Q statistics are established for the purpose of fault detection. Experiments on an industrial benchmark Tennessee Eastman process and a real blast furnace ironmaking process are carried out to demonstrate the superiority and effectiveness of the proposed method.

Genomics involving tens of thousands of genes is a complex system determining phenotype. An interesting and vital issue is that how to integrate highly sparse genetic genomics data with a mass of minor effects into prediction model for improving prediction power. We find that deep learning method can work well to extract features by transforming highly sparse dichotomous data to lower dimensional continuous data in a non-linear way. This idea may provide benefits in risk prediction based on genome-wide data associated e.g. integrating most of the information in the genotype data. Hence, we developed a multi-stage strategy to extract information from highly sparse binary genotype data and applied it for risk prediction. Specifically, we first reduced the number of biomarkers via a univariable regression model to a moderate size. Then a trainable auto-encoder was used to extract compact representations from the reduced data. Next, we performed a LASSO problem process over a grid of tuning parameter values to select the optimal combination of extracted features. Finally, we applied such feature combination to two prognostic models, and evaluated predictive effect of the models. The results of simulation studies and real data applying indicated that these highly compressed transformation features could better improve predictive performance and did not easily lead to over-fitting.

This study addressed the need for efficient data collection in micro-mobility using low-end hardware, focusing on developing a system capable of handling large real-time data streams. We proposed a novel data collection system that leverages incremental learning techniques, enabling AI-based services for micro-mobility passengers through smartphones. Our approach utilizes 3-axis acceleration sensors integrated into micro-mobility devices, collecting motion data in real-time. This data is processed through an incremental learning pre-processing server, employing the Broad Learning System algorithm, to ensure robust performance in noisy, real-world environments. The use of incremental learning is pivotal, allowing our system to adapt continuously to new data, enhancing the accuracy and relevance of the AI services provided to passengers. This research not only advances the technical capabilities in micro-mobility data handling but also has the potential to enhance the overall passenger experience through improved AI-based services.

In the era of digital advancements, the escalation of credit card fraud necessitates the development of robust and efficient fraud detection systems. This paper delves into the application of machine learning models, specifically focusing on ensemble methods, to enhance credit card fraud detection. Through an extensive review of existing literature, we identified limitations in current fraud detection technologies, including issues like data imbalance, concept drift, false positives/negatives, limited generalisability, and challenges in real-time processing. To address some of these shortcomings, we propose a novel ensemble model that integrates a Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Random Forest (RF), Bagging, and Boosting classifiers. This ensemble model tackles the dataset imbalance problem associated with most credit card datasets by implementing under-sampling and the Synthetic Over-sampling Technique (SMOTE) on some machine learning algorithms. The evaluation of the model utilises a dataset comprising transaction records from European credit card holders, providing a realistic scenario for assessment. The methodology of the proposed model encompasses data pre-processing, feature engineering, model selection, and evaluation, with Google Colab computational capabilities facilitating efficient model training and testing. Comparative analysis between the proposed ensemble model, traditional machine learning methods, and individual classifiers reveals the superior performance of the ensemble in mitigating challenges associated with credit card fraud detection. Across accuracy, precision, recall, and F1-score metrics, the ensemble outperforms existing models. This paper underscores the efficacy of ensemble methods as a valuable tool in the battle against fraudulent transactions. The findings presented lay the groundwork for future advancements in the development of more resilient and adaptive fraud detection systems, crucial as credit card fraud techniques continue to evolve.

While machine learning (ML) has been quite successful in the field of structural health monitoring (SHM), its practical implementation has been limited. This is because ML model training requires data containing a variety of distinct damaged instances captured from a real structure and the experimental generation of such data is challenging. One way to tackle this issue is by generating training data through numerical simulations. However, simulated data cannot capture the bias and variance of experimental uncertainty. To overcome this problem, this work proposes a deep learning-based domain transformation method for transforming simulated data to the experimental domain. Use of this technique has been demonstrated for debonding location and size prediction of stiffened panels using a vibration-based method. The results are satisfactory for both, debonding location and size prediction. This domain transformation method can be used in any field where experimental data for training machine learning models is scarce.

The novel coronavirus disease (COVID-19) pandemic is provoking a prevalent consequence on mental health because of less interaction among people, economic collapse, negativity, fear of losing jobs, and death of the near and dear ones. To express their mental state, people often are using social media as one of the preferred means. Due to reduced outdoor activities, people are spending more time on social media than usual and expressing their emotion of anxiety, fear, and depression. On a daily basis, about 2.5 quintillion bytes of data are generated on social media, analyzing this big data can become an excellent means to evaluate the effect of COVID-19 on mental health. In this work, we have analyzed data from Twitter microblog (tweets) to find out the effect of COVID-19 on peoples mental health with a special focus on depression. We propose a novel pipeline, based on recurrent neural network (in the form of long-short term memory or LSTM) and convolutional neural network, capable of identifying depressive tweets with an accuracy of 99.42%. Preprocessed using various natural language processing techniques, the aim was to find out depressive emotion from these tweets. Analyzing over 571 thousand tweets posted between October 2019 and May 2020 by 482 users, a significant rise in depressing tweets was observed between February and May of 2020, which indicates as an impact of the long ongoing COVID-19 pandemic situation.

In the extensive monitoring of maritime traffic, maritime management frequently encounters incomplete automatic identification system (AIS) data, particularly in critical static fields like vessel types. This shortfall presents substantial challenges in safety management, necessitating robust methods for data completion. Distinct from road traffic, where road width imposes constraints, maritime vessels of identical classifications often choose significantly divergent routes, deviating by several nautical miles. Sole dependence on trajectory clustering is insufficient. Addressing this issue, we segments the maritime area into spatiotemporal grids, transforming vessel paths into sequences of grid encodings. Utilizing natural language processing techniques, we integrate the Word2vec word embedding approach with our novel biLSTM self-attention chunk-max pooling net (biSAMNet) model, enhancing the classification of vessel trajectories. This method is crucial for determining static details like vessel types. Employing the Taiwan Strait as a case study and benchmarking against CNN, RNN, and methods based on the attention mechanism, our findings underscore our model’s superiority. The biSAMNet achieves an impressive vessel classification F1 score of 0.94, using only 5-dimensional word embeddings, highlighting the effectiveness of Word2vec pre-trained embedding layers. This research introduces a novel paradigm for processing vessel trajectory data, greatly enhancing the accuracy of filling in incomplete vessel type details.

The research proposes a self-directed paradigm with self-design using the existing technology with LMS modules, discussions, and self-tests. The research establishes its criteria for assurance of teaching and learning known as 7x2C for the compliance of content, and appearance along with other criteria for the analysis. The above technology of web and LMS is tested for a business course that is technology-oriented in creating an experiential learning system for Covid-19 awareness. Similarly, among other ideas, an environment for educating learners about diabetes and obesity has been designed. The project is known as Sunchoke with its theme of Grow, Eat, and Heal. The educators can use their technology and apply self-directed pedagogy to transform this approach to their customized ways by using their content and rubrics.

Fabry-Pérto (FP) cavity is the essential component of ultra-stable laser (USL) for gravitational wave detection, which couples multi-physic (Optical/Thermal/Mechanics) fields and requires ultra-high precision. To satisfy the requirements of precise and efficient design, a multi-physic coupling optimization method for fixing cubic FP cavity based on data learning is proposed. A multi-physic model for the cubic FP cavity is established and the performance is obtained by finite element analysis. The key performance indices (V, wF, wF) and key design variables (d, l, F) are determined considering the features of the FP cavity. The 49 sets of data by orthogonal experiment are acquired for the establishment and comparison of different data learning models (NN, RSF, KRG). The result turns out that the neural network has the best performance. Based on NSGA-II, the Pareto optimal front is obtained and the optimal combination of design variables is finally determined as {5,32,250}. The performance after optimization has proven to be a great improvement, of which the displacement under the fixing force and vibration test are decreased by more than 60%. The optimization strategy can not only help the design of the FP cavity but also enlighten other optimization fields.

Development of supervised AI algorithms requires a large amount of labeled images. Image labelling is both time-consuming and expensive. Therefore, we explored the value of e-learning derived annotations for AI algorithm development in medical imaging. Methods We have developed an e-learning platform that involves image-based single click labelling as part of the educational learning process. Ten radiology residents, as part of their residency training, trained the recognition of pneumothorax on 1161 chest X-rays in posterior-anterior projection. Using this data, multiple AI algorithms for detecting pneumothorax were developed. Classification and localization performance of the models was tested on an independent internal testing dataset and on the public NIH ChestX-ray14 dataset. Results The AI models F1 scores on the internal and the NIH dataset were 0.87 and 0.44, respectively. Sensitivity was 0.85 and 0.80 for classification and specificity 0.96 and 0.48 for classification. F1 scores were 0.72 and 0.66, sensitivity 0.72 and 0.72. False positive rate was 0.36 and 0.32 for localisation. Conclusion Our results demonstrated that e-learning derived annotations are a valuable data source for algorithm development. Further work is needed to include additional parameters such as user performance, consensus of diagnosis, and quality control in the development pipeline.