This paper describes a memory model with 2 levels of information. The lower-level stores source data, is Markov-like and unweighted. Then an upper-level ontology is created from a further 3 phases of aggregating source information, by transposing from an ensemble to a hierarchy at each level. The ontology is useful for search processes and the aggregating process transposes the information from horizontal set-based sequences to more vertical typed-based clusters. The base memory is essentially neutral, where any weighted constraints or preferences should be sent by the calling module. This therefore allows different weight sets to be imposed on the same linking structure. The success of the ontology typing is open to interpretation, but the author would suggest that when clustering text, the result was types based more on use and context, for example, 'linking' with 'structure' or 'provide' with 'web,' for a document describing distributed service-based networks. This allows the system to economise over symbol use, where links to related symbols will be clustered together. The author then conjectures that a third level would be more neural in nature and would include functions or operations to be performed on the data, along with related memory information.

Evolutionary Machine Learning (EML) has demonstrated its effectiveness in modeling various real-world application domains, including the dynamics of diseases and the creation of predictive models across diverse fields. However, the challenge of data imbalance adversely impacts the performance of EML models. While strategies involving misclassification costs or cost-sensitive learning exist to address this issue, their integration into EML has yet to be thoroughly explored. This paper aims to fill this gap by investigating the application of cost-sensitive EML for intelligent health risk prediction. More specifically, this paper delves into the utilization of cost-sensitive genetic programming (CSGP) and cost-sensitive genetic algorithms (CSGA), which are among the most widely used algorithms in the EML category. The study entails training CSGP and CSGA to predict the early onset of three chronic conditions (hypertension, diabetes, and fatty acid-related disorders) through the development of an EML model. Subsequently, the model's performance is evaluated through standard fitness functions such as sensitivity, specificity, and F1-score. From the experimental results, EML models demonstrate promising results in identifying the occurrence of chronic diseases. Further optimization of the cost-sensitive EML model holds the potential for enhanced outcomes in modeling medical problems.

Vehicular traffic in urban areas continues to be a constant challenge today, due to the continuous population growth and the increase of vehicles in circulation. Vehicular congestion, as a recurrent problem, generates a negative impact on urban mobility and citizens’ quality of life. It is hypothesized that a dynamic clustering method of vehicle trajectory data can provide an accurate and up-to-date representation of real-time traffic behavior. To evaluate this hypothesis, data were collected from three different cities: San Francisco, Rome and Guayaquil. A dynamic clustering algorithm was applied to identify traffic congestion patterns and an indicator was applied to identify and evaluate the congestion conditions of the areas. The results show a high precision and recall in congestion classification compared to a method based on static cells.

The incorporation of renewable energy systems in the world energy system has been steadily increasing during the last years. In terms of the building sector, the usual consumers are becoming increasingly prosumers, and the trend is that communities of energy, whose households share produced electricity, will increase in number in the future. Another observed tendency is that the aggregator (the entity that manages the community) trades the net community energy in public energy markets. To accomplish economically good transactions, accurate and reliable forecasts of the day-ahead net energy community must be available. These can be obtained using an ensemble of multi-step shallow artificial neural networks, with prediction intervals obtained by the covariance algorithm. Using real data obtained by a small energy community of four houses located in the South region of Portugal, one can verify that the deterministic and probabilistic performance of the proposed approach is at least similar, typically better than using complex, deep models.

This research explores the transformation of learning methodologies from traditional to modern approaches, focusing on the development of a gamified educational application with an Artificial Intelligence (AI) mentor for secondary school graduates preparing for Croatian state graduation exams. Traditionally, education has been marked by a passive student role and teacher-centric transmission of knowledge. In contrast, modern learning tools, such as gamified applications, introduce interactive and engaging elements. The purpose of this study is to identify user-centric learning approaches through a comprehensive survey of graduating students, assessing their readiness for exams, learning styles, and preferences. The research explores issues such as financial burdens on parents for preparatory courses and copyright concerns with external educational materials. In summary, the learning application should prioritise sustained knowledge retention, motivation, and stress reduction, offering an alternative to external preparatory courses. Conclusions underscore the importance of fostering an engaging learning environment, incorporating teacher-like elements, and aligning with user expectations. As digital technology positively impacts the learning experience, understanding user preferences is crucial in designing educational apps that cater to learners' needs. The study provides insights for developing a gamified learning application that not only prepares students for exams but also enhances long-term knowledge acquisition and engagement.

Query reformulation is a fundamental task in information retrieval systems aimed at improving the accuracy and relevance of search results. This abstract presents a query reformulation approach that utilizes a domain-specific ontology to enhance the effectiveness of query reformulation. The use of ontology enables a more precise understanding of the domain and its associated concepts, leading to more accurate query interpretation and reformulation. The proposed approach consists of two main steps: ontology-based query interpretation and ontology-guided query expansion. The proposed approach use Power thesaurus tool that perform sentiment analysis for suggestion of the synonyms of the query words to improve the accuracy of the retrieved results. The proposed methodology gives high accuracy as compared to the previous approaches that used Wordnet for the suggestions of synonyms. The proposed approach demonstrates the potential of utilizing ontologies in query reformulation, offering a more accurate and context-aware search experience in domain-specific information retrieval systems.

The use of Artificial Intelligence (AI) applications in a growing number of domains in the latest years has put into focus Ethical Legal and Societal Aspects (ELSA) of these technologies and the relevant challenges they pose. In this paper, we propose an ELSA Curriculum for Data Scientists aiming to raise awareness about ELSA challenges in their work, provide them with a common language with the relevant domain experts to cooperate to find appropriate solutions, and finally, incorporate ELSA in the data science workflow and not be seen as an impediment or a superfluous artifact, rather than an integral part of the Data Science Project Lifecycle. The proposed curriculum uses the CRISP-DM model as a backbone to define a vertical partition, expressed in modules corresponding to the CRISP-DM phases. The horizontal partition includes knowledge units belonging to three strands that run through the phases, namely Ethical and Societal, Legal, and Technical Rendering Knowledge Units (KUs). In addition to the detailed description of the aforementioned KUs, we also discuss the implementation, issues such as duration, form, and evaluation of participants, as well as the variance of the knowledge level and needs of the target audience.

This review explores the integration of Virtual and Mixed Reality in medical training. The advent of immersive technologies has revolutionized traditional methodologies, offering immersive, interactive, and flexible learning environments. This study aims to assess the impact, efficacy, and potential of Extended Reality (XR) in medical education. Methods included a comprehensive analysis of current Metaverse’s applications in various medical fields, including surgery, diagnostics, and patient care. Results indicate a significant enhancement in learning outcomes, practical skills, and student engagement. Conclusions suggest that Virtual Reality is a valuable tool for medical training, providing realistic simulations and improving both theoretical knowledge and practical skills. The intersection of virtual reality and cognitive neuroscience presents an exciting frontier for understanding the human brain's response to immersive environments. This review explores how VR impacts learning, with a particular focus on the activation and engagement of the prefrontal cortex during immersive experiences. It delves into various studies that highlight the transformative potential of VR in educational settings and therapeutic applications, demonstrating how VR can facilitate deeper and more effective learning experiences. The review will present a dedicated section on the current methodologies used to analyze the brain's response while immersed in a virtual environment.

Convolutional neural networks (CNNs) have proven to be a very efficient class of machine learning (ML) architectures for handling multidimensional data by maintaining data locality, especially in the field of computer vision. Data pooling, a major component of CNNs, plays a crucial role for extracting important features of the input data and downsampling its dimensionality. Multidimensional pooling, however, is not efficiently implemented in existing ML algorithms. In particular, quantum machine learning (QML) algorithms have a tendency to ignore data locality for higher dimensions by representing/flattening multidimensional data as simple one-dimensional data. In this work, we propose using the quantum Haar transform (QHT) and quantum partial measurement for performing generalized pooling operations on multidimensional data. We present the corresponding decoherence-optimized quantum circuits for the proposed techniques along with their theoretical circuit depth analysis. Our experimental work was conducted using multidimensional data, ranging from 1-D audio data, to 2-D image data, to 3-D hyperspectral data, to demonstrate the scalability of the proposed methods. In our experiments, we utilized both noisy and noise-free quantum simulations on a state-of-the-art quantum simulator from IBM Quantum. We also show the efficiency of our proposed techniques for multidimensional data by reporting the fidelity of results.

Startups are playing increasingly influential roles in the technology sector of the world. India has been a rapidly growing economy and hosts over one hundred thousand total startups. Investors have been increasingly investing in the Indian technology sector. However, the Indian startup ecosystem is different to the American startup ecosystem and requires a separate analysis to determine important influences for their success. Through this research with responsible machine learning, entrepreneurs will be empowered to better understand how to successfully raise their company. I collected data from Crunchbase and defined a successful startup as one who has acquired another company, was acquired, or went public. I used Random Forest, XGBoost, LightGBM, and CatBoost to predict the success of the startups. To determine the most important factors, I used the feature importance tools provided by the models. I compared these results and found that the time taken between the founding and first funding of the company, commonly referred to as seed lag, was the most pivotal factor to every model’s prediction of success.

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.

Multispectral and 3D LiDAR remote sensing data sources are valuable tools for characterizing the 3D vegetation structure and thus understanding the relationship between forest structure, biodiversity and microclimate. This study focuses on mapping riparian forest species in the canopy strata using a fusion of Airborne LiDAR data and multispectral multi-sources and multi-resolution satellite imagery; Sentinel-2 and Pleiades at tree level. The idea is to assess the contribution of each data source in the tree species classification at the considered level. The data fusion was processed at feature-level and decision level. At feature level, LiDAR 2D attributes were derived and combined with vegetation indices. At decision level, LiDAR data was used for 3D tree crown delimitation providing unique trees or groups of trees that are used as a support for the species classification. Data augmentation techniques were used to improve the training process. Best results were obtained by the fusion of Sentinel-2 time series and LiDAR data with a Kappa of 0.66 thanks to red-edge based indices that better discriminate vegetation species and the temporal resolution of Sentinel-2 images that allows monitoring the phenological stages helping to discriminate the species.

This paper introduces the AltumView Sentinare smart activity sensor for senior care and patient monitoring. The sensor uses an AI chip and deep learning algorithms to monitor the activity of people, collect activity statistics, and notify caregivers when emergencies such as falls are detected. To protect privacy, only skeleton (stick figure) animations are transmitted instead of videos. The sensor is highly affordable, accessible, and versatile. It was a CES 2021 Innovation Award Honoree, and has been selected by Amazon as one of only three fall detection devices integrated into its Alexa Together urgent response service, and has received very positive reviews from Amazon customers. It has also been used in different senior care settings in about ten different countries. The paper presents the main features of the system, the evidences and lessons learned from its practical applications, and future directions.

This research delves into the realm of predictive modeling within artificial intelligence (AI) and introduces an innovative ensemble approach designed to enhance prognostic precision. The proposed methodology, characterized by its ebullient nature, leverages the collective wisdom of diverse predictive models, synthesizing a comprehensive and robust framework. The pursuit of parsimony are paramount, underscoring the need for efficient model architectures that balance predictive acuracy and computational efficiency. Our study meticulously explores the amalgamation of diverse algorithms, encapsulating the intrinsic complexity of real-world phenomena. We navigate the intricate landscape of feature selection, algorithmic diversity, and model combination, fostering a holistic understanding of the prognostic process. The ebullient ensemble emerges as a dynamic solution, harnessing the synergy of constituent models while mitigating the risk of overfitting. To validate the efficacy of our approach, extensive experiments are conducted on diverse datasets, spanning domains such as healthcare, finance, and natural language processing. Comparative analyses demonstrate the superior predictive performance of our ensemble model in comparison to individual algorithms and traditional ensemble methods. Additionally, we delve into interpretability aspects, elucidating the underlying mechanisms that contribute to the model's prognostic prowess. Our research advances the frontier of predictive modeling in AI, offering a nuanced and sophisticated paradigm for prognostication. The ebullient ensemble approach stands as a testament to the synergy achievable through the amalgamation of diverse predictive models, providing a pragmatic and parsimonious avenue for enhanced prognostic acuracy in artificial intelligence applications.

The precise determination of the State-of-Health (SOH) of lithium-ion batteries is critical in the domain of battery management systems. The proposed model in this research paper emulates any deep learning or machine learning model by utilizing a Look Up Table (LUT) memory to store all activation inputs and their corresponding outputs. The operation that follows the completion of training is referred to as the LUT memory preparation procedure. The lookup operation performed on this method simply substitutes for the inference process. This is achieved by discretizing the input data and features before binarizing them. The term for the aforementioned operation is the LUT inference method. The procedure was evaluated in this study using two distinct neural network architectures: a bidirectional long short-term memory (LSTM) architecture and a standard fully connected neural network (FCNN). It is anticipated that considerably greater efficiency and velocity will be achieved during the inference procedure when the pre-trained deep neural network architecture is inferred directly. The principal aim of this research is to construct a lookup table that effectively establishes correlations between the SOH of lithium-ion batteries and ensures a degree of imprecision that is tolerable. According to the results obtained from the NASA PCoE lithium-ion battery dataset, the proposed methodology exhibits performance that is largely comparable to that of the initial machine learning models. Utilising the error assessment metrics RMSE, MAE, and (MAPE), the accuracy of SOH prediction has been quantitatively evaluated. The indicators mentioned above demonstrate a significant degree of accuracy when predicting SOH.

Survival models have become popular for credit risk estimation. Most current credit risk survival models use an underlying linear model. This is beneficial in terms of interpretability but is restrictive for real-life applications since it cannot discover hidden nonlinearities and interactions within the data. This study uses discrete time survival models with embedded neural networks as estimators of time to default. This provides flexibility to express nonlinearities and interactions between variables, and hence allows for models with better overall model fit. Additionally, the neural networks are used to estimate Age-Period-Cohort (APC) models so that default risk can be decomposed into time components for loan age (maturity), origination (vintage) and environment (e.g., economic, operational and social effects). These can be built as general models, or as local APC models for specific customer segments. The local APC models reveal special conditions for different customer groups. The corresponding APC identification problem is solved by a combination of regularization and fitting the decomposed environment time risk component to macroeconomic data, since the environmental risk is expected to have a strong relationship with macroeconomic conditions. Our approach is shown to be effective when tested on a large publicly available US mortgage data set. This novel framework can be adapted by practitioners in the financial industry to improve modelling, estimation and assessment of credit risk.

UAVs are becoming a crucial tool in modern warfare, primarily due to their cost-effectiveness, risk reduction, and ability to perform a wider range of activities. The use of autonomous UAVs to conduct strike missions against highly valuable targets is the focus of this research. Due to developments in ledger technology, smart contracts, and machine learning, such activities formerly carried out by professionals or remotely flown UAVs are now feasible. Our study provides the first in-depth analysis of challenges and potential solutions for successful implementation of an autonomous UAV mission.

Although the behavioral mechanism studying is a difficult and complex task, it can produce an important systems optimization impact. In this work, we use, in this paper, inference system approach to represent reasoning mechanism and the operations of dynamic systems to extract mental representations from traveling salesman and convert them into cognitive structures. For this we develop an extraction automatic method to create knowledge bases and, later, data are stocked into structure based on transition maps and the performances of these created maps get improved through combining the reinforcement learning thus augment traveller's deciding capacity from historical data. These transition maps help to find best actions for obtaining useful new policies. Generated intermediate transition maps are gathered to give a global map called main map whose advantage is to improve the learning process. The main idea of this approach is to improve learning capability by using a reinforcement learning technique as exploration and exploitation strategy of the metaheuristique GRASP method and the use of the fuzze logical rules mechanism allowing concepts model to have more variability of states. The results obtained after simulation as presented in this paper are very encouraging.

Accurate medical image segmentation is paramount for precise diagnosis and treatment in modern healthcare. This research presents a comprehensive study on the efficacy of Particle Swarm Optimization (PSO) combined with Histogram Equalization (HE) preprocessing for medical image segmentation, focusing on Lung CT-Scan and Chest X-ray datasets. Best Cost values reveal the PSO algorithm’s performance, with HE preprocessing demonstrating significant stabilization and enhanced convergence, particularly in complex Lung CT-Scan images. Evaluation metrics, including Accuracy, Precision, Recall, F-Score, Specificity, Dice, and Jaccard, show substantial improvements with HE preprocessing, emphasizing its impact on segmentation accuracy. Comparative analyses against alternative methods, such as Otsu, Watershed, and K-means, confirm the competitiveness of the PSO-HE approach, especially in Chest X-ray images. The study also underscores the positive influence of preprocessing on image clarity and precision. These findings highlight the promise of the PSO-HE approach in advancing the accuracy and reliability of medical image segmentation, paving the way for further research and method integration to enhance this critical healthcare application.

Cognitive Warfare (CogWar) is another form of warfare where the goal is to exploit cognition facets to disrupt, undermine, influence or modify human decisions. Famous examples include “Operation Gidlock” meddling with the US 2016 election, and CogWar campaigns assisting the annexation of Crimea. As Large Language Model(LLM)-based applications expand, reaching more users and their cognitive states, stealth Malicious Cognitive Behavioral Tactics (MCBT) can be embedded in trusted application sessions to systematically profile, then alter, each individual’s cognition over time. Successful MCBT can create more cases like Snowden, mass shootings, or January6. This short-form paper discusses a novel MCBT threat model, a novel kill chain, and an on-going prototype for demonstration.

Unsupervised physics-informed deep learning can be used to solve computational physics problems by training neural networks to satisfy the underlying equations and boundary conditions without labeled data. Parameters such as network architecture and training method determine the training success. However, the best choice is unknown a priori as it is case specific. Here, we investigated network shapes, sizes, and types for unsupervised physics-informed deep learning of the two-dimensional Reynolds-averaged flow around cylinders. We trained mixed-variable networks and compared them to traditional models. Several network architectures with different shape factors and sizes were evaluated. The models were trained to solve the Reynolds-averaged Navier-Stokes equations incorporating Prandtl's mixing length turbulence model. No training data were deployed to train the models. The superiority of the mixed-variable approach was confirmed for the investigated high Reynolds number flow. The mixed-variable models were sensitive to the network shape. For the two cylinders, differently deep networks showed superior performance. The best fitting models were able to capture important flow phenomena such as stagnation regions, boundary layers, flow separation, and recirculation. We also encountered difficulties when predicting high Reynolds number flows without training data

With the development of electricity spot markets, accurate electricity load forecasting enables power generation companies to supply the right amount of electricity, Significantly avoid power waste. As a result, time series forecasting in the field of power can bring significant benefits. Previously, the Informer model successfully introduced the Transformer into long time series forecasting(LTSF) by proposing the ProbSparse self-attention mechanism, which solved the inherent problem of high memory complexity in self-attention. Recent research has further demonstrated the potential of the self-attention for mining complex dependencies. However, the limited amount of historical data has become one of the main challenges in applying deep learning techniques to power LSTF tasks. Previous reserches often add a large number of time covariates to provide more information. In this paper, to address this issue, (i) we design a simple but effective Season-aware Block to enhance the model’s ability to mine artificial prior information in temporal covariates; (ii) we conduct experiments using the provincial power data of Zhejiang Province, China, from 2019 to 2022, and our model outperforms other models, achieving a 19 percent MSE relative improvement; (iii) we conduct ablation experiments to assess the efficacy of the Season-aware Block in extracting temporal periodic features. Furthermore, we elucidate the underlying reasons for the effectiveness of both the self-attention mechanism and the Season-aware Block through visualization experiments.

In this study, we introduce a neural-learning based approach for the early detection of Parkinson Disease using DaT SPECT imaging, a promising technique in neurodegenerative disease diagnosis. Leveraging the advanced capabilities of deep learning, we developed a convolutional neural network model based on the Inception V3 architecture, which has been enhanced by modulating custom layers to fine-tune the model specifically for the analysis of DaT SPECT images. The imaging dataset was obtained through the Parkinson's Progression Markers Initiative (PPMI), and the model was trained to classify whether a patient had Parkinson Disease through a binary classification of their DaT SPECT Scan using the fine-tuned InceptionV3 model. This specialized adaptation allows for more accurate and nuanced interpretation of the scans, a crucial factor in early and reliable diagnosis of Parkinsonian syndromes. Our model underwent rigorous hyperparameter tuning, focusing on learning rates, the number of neurons in the dense layer, and dropout rates, to ensure optimal performance. Remarkably, the proposed model demonstrated a significant improvement in accuracy, achieving 74.65%, compared to a standard Sequential Model, which only reached an accuracy of 51.41%. Diagnosis of Parkinson Disease through DaT SPECT Scans is a promising approach not only as a more effective diagnostic tool but also as a beacon for future research in the application of deep learning to medical imaging for neurodegenerative diseases. By building layers on the InceptionV3 Architecture, we achieve a higher accuracy on DaT SPECT Scan diagnosis and pave the way for automated diagnosis of patients with Parkinson Disease, allowing for easier medical accessibility and paving the way for advancements in early diagnosis and personalized medicine strategies for Parkinson's disease and related disorders.

There are many applications of anomaly detection in IoT domain. IoT technology consists of large number of interconnecting digital devices not only generating huge data continuously but also making real-time computations. Since IoT devices are highly exposed due to Internet, they frequently meet with the challenges of illegitimate accesses in the form of intrusions, anomaly, fraud, etc. Identifying these illegitimate accesses in IoT domain can be an exciting research problem. In numerous applications fuzzy clustering and rough set theory have been successfully employed. As the data generated in IoT domains are high-dimensional, the clustering methods used for lower dimensional data cannot be applied efficiently. In this article, mixed approaches consisting of nano topology and fuzzy clustering techniques are proposed for anomaly detection. First of all, the nano topology is generated to find lower dimensional apace and then a couple of well-known fuzzy clustering techniques are employed on it for the efficient anomaly detection. The effectiveness of the proposed approaches is evaluated using time-complexity analysis, experimental studies with a synthetic dataset and a real-life dataset along with comparative studies with traditional fuzzy clustering approaches namely fuzzy c-means clustering (FCM) algorithm, Gustafson-Kessel (GK) Algorithm, Gath-Geva (GG) Algorithm, Mahalanobis Distance based Fuzzy C-Means algorithm (M-FCM), and Common Mahalanobis Distance based Fuzzy C-Means algorithm (CM-FCM). Experimentally, it has been found that the proposed approaches outperform the aforesaid algorithms in terms of detection rates, accuracy rates, false alarm rates and computational times.

Along with the rapid development and wide application of Artificial Intelligence (AI), discussions regarding AI Sustainability have emerged. Employing a systematic mapping study, this paper reviews research about AI Sustainability from diverse fields and viewpoints, including three dimensions economic, social, and environmental. Our analysis indicates a transformative shift in research focus, evolving from fragmented approaches to a more holistic understanding of AI Sustainability. This maturation is reflected in the interplay between multiple dimensions and the integration of both "AI as a tool for sustainability" and "Sustainability of AI" perspectives. As the field continues to mature, researchers should embrace a comprehensive and unbiased exploration of AI Sustainability across all dimensions and approaches, ensuring a well-rounded foundation for future advancements in this rapidly evolving domain.

Fall detection systems play a crucial role in addressing the significant health concern of elderly falls, a leading cause of health deterioration and mortality. As the aging population grows and life expectancy increases, the development of accessible tools becomes vital for predicting and preventing falls, offering a practical and widely applicable solution in contrast to costly and expertise-dependent assessment tools. In contrast, due to the formidable challenges encountered, a comprehensive investigation into the comparative performance of standard ML models within this field still needs to be explored. This paper proposes a standard pipeline for pre-processing, training, and evaluating ML models for fall detection on the SisFall dataset. We conducted extensive experiments to evaluate the performance of various ML models for fall detection. The results validate the efficiency of the deep model in identifying the time windows in which a fall occurred. Among the deep models, the architecture, including a combination of convolutional neural networks and fully connected layers, outperforms the others by macro-averaged Precision, macro-averaged Recall, and macro-averaged F1-Score of 87.03\%, 86.83\%, and 86.93\%, respectively.

In areas where conventional fundus cameras are logistically unaffordable, the smartphone-based approach is considered as a promising future of glaucoma screening due to their affordability and ease-of-use. Optic disc localization is a critical stage during glaucoma screenings. For fundus images captured with standard fundus cameras, the majority of the models available out there can locate the optic disc with satisfactory performance. However, for images that are captured with smartphone-based fundus cameras, the inherent noise and lower quality of the fundus image makes it difficult for models to detect the optic disc region. In this study, we have proposed to utilize YOLOv8n for optic disc localization due to the model’s cutting-edge performance in diverse tasks and its lightweight nature. We have used a public dataset which has 2000 low quality fundus images that are captured using a smartphone-based fundoscopy device. From these, 60% of the data was used for fine-tuning the model, and 25% for testing. By using a confidence of 50% set as threshold, the model was able to detect the optic disc successfully on over 97% of test images with intersection over union of above 0.85. These results highlight the potential of the lightweight YOLOv8n model for deployment on resource-constrained environments, offering a promising performance on accurately localizing the OD and enhancing the feasibility of affordable glaucoma screening on smartphones.

Continual learning, a crucial facet of machine learning, involves the perpetual acquisition of valuable insights from incoming data, sans the necessity for full dataset access. Esteemed as a fundamental goal in artificial intelligence, continual learning grapples with an enduring challenge—catastrophic forgetting. A proficient model must exhibit adaptability for new data assimilation and robustness to retain existing knowledge. Class-incremental learning (CIL) aids the gradual integration of knowledge from newly introduced classes, forming a universal classifier. However, directly training the model with fresh class instances triggers a problem—forgetting distinguishing features of prior classes, causing a performance decline. Addressing such issues in machine learning, this survey aims to delineate significant challenges, outcomes, and recent advancements, including our contributions to CIL techniques, especially in image classification.

This study presents a novel approach to enhance Large Language Models (LLMs) like Alpaca by dynamically integrating real-time information. This method addresses the issue of content hallucination and data relevancy by automatically collecting and integrating current data from credible sources into model prompts. Experiments show a significant improvement in accuracy and a decrease in content hallucination, with a manageable increase in response time. The research underscores the potential of real-time data integration in making LLMs more accurate and contextually relevant, setting a foundation for future advancements in dynamic data processing in AI.

Pathological Brain Detection (PBD) is a crucial field aimed at identifying and diagnosing structural or functional abnormalities in the brain related to neurological, neurodegenerative, or psychiatric disorders. This detection process typically employs medical imaging technologies such as MRI, CT, or PET scans, along with neurological evaluations, blood tests, and other diagnostic tools. Early detection is essential for effective treatment, improving prognosis, and enhancing the quality of life for affected individuals. Each common computational method provides unique insights into brain pathologies. Despite significant advancements in technology and methodology, PBD faces challenges such as variability in brain anatomy, the complexity of disorders, data quality, standardization issues, interpretability of models, and ethical concerns. Addressing these challenges necessitates collaboration among researchers, clinicians, and policymakers to develop robust and ethical methods for improving the detection and diagnosis of brain disorders.

In recent years, advancements in rooftop solar panel technology have sparked a revolution in the electricity markets. This has given rise to a new concept of energy exchange - the ability for consumers and producers to trade localized energy. This concept has been made possible by the emergence of blockchain technology, which has gained significant traction in the energy markets. Its unique ability to facilitate peer-to-peer (P2P) energy transactions has made it a promising solution for the trilemma of scalability, security, and decentralization. However, while blockchain technology has shown great potential, it is still in its early stages of development and has yet to reach its full potential. To fully understand the potential of blockchain technology in P2P energy trading, it is important to explore the trilemma in depth. This study proposes a blockchain-based scalability solution, with a focus on P2P energy trade. This strategy is supported by empirical modeling, utilizing data gathered from a trial case study. The results of this study demonstrate that the suggested technique outperforms base layer models in terms of scalability, while still maintaining the essential elements of security and decentralization. This proposed solution not only has the potential to revolutionize the energy markets, but also has broader implications. By providing a more secure and decentralized platform for P2P energy trading, it has the potential to address issues of energy distribution inequality and promote the adoption of renewable energy. With blockchain technology, individuals and communities have the opportunity to take control of their energy usage and production, reducing their reliance on traditional centralized energy systems. This not only has the potential to lower energy costs for consumers, but also contributes to the overall goal of reducing carbon emissions and mitigating the effects of climate change. The combination of rooftop solar panel technology and blockchain applications has the potential to create a significant shift towards decentralized and sustainable energy systems. This will not only benefit individuals and communities, but also have a positive impact on the environment and the global energy market as a whole. However, for this transition to occur, it is crucial that governments and energy companies continue to support and invest in these advancements.

The phenomenon of missing data can be seen everywhere in reality. Most typical attribute reduction models are only suitable for complete systems. But for incomplete systems, we can’t obtain the effective reduction rules. Even if there are a few reduction approaches, the classification accuracy of their reduction sets still needs to be improved. In order to overcome these shortcomings, this paper firstly defines the similarities of intra-cluster objects and inter-cluster objects based on the tolerance principle and the mechanism of knowledge granularity. Secondly, attributes are selected on the principle that the similarity of inter-cluster objects is small and the similarity of intra-cluster objects is large, and then the knowledge granularity attribute model is proposed under the background of clustering; Then, the IKAR algorithm program is designed. Finally, a series of comparative experiments about reduction size, running time and classification accuracy are conducted with twelve UCI data sets to evaluate the performance of IKAR algorithms, then the stability of Friedman test and Bonferroni-Dunn test are conducted. The experimental results indicate that the proposed algorithms are efficient and feasible.

The paper aims to develop and test an algorithm for classifying Russian-language text reviews of patients’ experiences with medical facilities and physicians, extracted from social media. This is motivated by the limitations of conventional methods of surveying consumers to assess their satisfaction with the quality of services, which are being replaced by automatic processing of text data from social media. This approach enables to get more objective results due to the increased representativeness and independence of the sample of service consumers. The authors have tested machine learning methods using various neural network architectures. A hybrid method was developed to classify text reviews of medical facilities posted by patients on the two most popular physician review websites in Russia. Overall, more than 60,000 reviews were analysed. The main results are as follows: 1) the classification algorithm developed by the authors has a high efficiency, the best result being achieved by the GRU-based architecture (val_accuracy = 0.9271); 2) applying the named entity search method to text messages following their partitioning improved the classification efficiency for each of the classifiers based on artificial neural networks. To further enhance the classification quality, reviews need to be semantically partitioned by target and sentiment and the resulting fragments need to be analysed separately.

Recognizing trust as a pivotal element for success within Human-Robot Collaboration (HRC) environments, this article examines its nature, exploring the different dimensions of trust, analysing the factors affecting each of them, and proposing alternatives for trust measurement. To do so, we designed an experimental procedure involving 50 participants interacting with a modified 'Inspector game' while we monitor their brain, electrodermal, respiratory, and ocular activities. This procedure allowed us to map dispositional (static individual baseline) and learned (dynamic, based on prior interactions) dimensions of trust considering both demographic and psychophysiological aspects. Our findings challenge traditional assumptions regarding the dispositional dimension of trust and establish clear evidence that the first interactions are critical for the trust-building process and the temporal evolution of trust. By identifying more significant psychophysiological features for trust detection and underscoring the importance of individualized trust assessment, this research contributes to understanding the nature of trust in HRC. Such insights are crucial for enabling more seamless human-robot interaction in collaborative environments.

AI applications are transforming the technology and digital sector, seamlessly integrating into the daily interactions of individuals with the digital world. Key entry points to the internet, such as search engines, are now complemented by recommendation systems that curate content on social networks, streaming platforms, and online marketplaces. Additionally, virtual assistants have become ubiquitous across various platforms. While these groundbreaking changes are gradually extending to several industries, there remains significant untapped potential for the application of artificial intelligence in diverse domains. The renewable energy sector, in particular, continually reinvents itself through research and innovation to meet the escalating demand. Positioned at the forefront of technological evolution, this industry is well-placed to effectively adopt and tailor the innovations of artificial intelligence to its specific needs. Presently, both emerging and established tools leverage elements of AI to enhance processes and introduce functionalities that were previously unattainable. If the current trajectory persists, AI-enhanced tools have the potential to assume a role as crucial to the future of engineering as computer-aided design (CAD) software is in contemporary times.

Process mining offers significant services in the development of digital threads and digital twin applications for healthcare. One of the sub-emerging case studies is the use of patients’ location data in process mining analyses. While the streamlining of published works is focused on introducing process discovery algorithms, there is a necessity to address challenges beyond that. Literature analysis indicates that explainability, reasoning, and characterizing the root causes of process drifts in healthcare processes is an important but overlooked challenge. In addition, incorporating domain-specific knowledge into process discovery could be a significant contribution to process mining literature. To address this, we present the DIAG approach that consists of a meta-model and an algorithm that enables diagnosing in process mining through an ontology-driven approach. With DIAG, we modeled the healthcare semantics in a process mining application and diagnosed the causes of drifts in patients’ pathways. We performed an experiment in a hospital living lab to examine the effectiveness of the DIAG approach.

In this paper, we introduce Syntax-Enhanced Transformer Model (SET), a groundbreaking approach in the realm of Transformer-based language modeling that seeks to redefine the boundaries of linguistic analysis and comprehension. SET innovatively combine (i) the well-established high-level performance, scalability, and adaptability of traditional Transformers with (ii) a sophisticated analysis of syntactic structures. This synergy is enabled by a novel attention mechanism tailored to parse syntactic nuances and a deterministic process adept at transforming linearized parse trees into meaningful linguistic representations. Our comprehensive experiments reveal that SET significantly advance the field by surpassing existing benchmarks in sentence-level language modeling perplexity. They exhibit exceptional proficiency in tasks that require an acute awareness of syntax, setting new standards for language models in understanding complex linguistic structures. Furthermore, SET demonstrate an enhanced capability to grasp nuanced linguistic patterns that have traditionally been challenging for standard Transformer models. However, our studies also uncover a unique aspect of SET: while they excel in sentence-level tasks, their representation of sentences as singular vectors—owing to the syntactic composition constraints intrinsic to their design—introduces certain limitations in document-level language modeling. This observation points to an intriguing area for future exploration; it suggests the potential need for an alternative or complementary memory mechanism within Transformer models, one that functions independently from, yet in harmony with, syntactic structures. Such a mechanism could be pivotal in enhancing the model's ability to comprehend and process long-form texts effectively. In conclusion, SET mark a significant stride in the journey towards more sophisticated, syntax-aware language models. They offer promising insights into the integration of deep linguistic knowledge with cutting-edge machine learning techniques, potentially opening doors to a new era of natural language understanding and processing.

The author previously reported an experimental result of evolutionary reinforcement learning of neural network controllers for the pendulum task. In the previous work, a conventional multilayer perceptron was employed in which connection weights were real numbers. In this study, the author experimentally applies an evolutionary algorithm to the reinforcement training of binary neural networks. In both studies, the same task and the same evolutionary algorithm are utilized, i.e. the pendulum control problem and Evolution Strategy respectively. The only differences lie in the memory size per connection weight and the model size of the neural network. The findings from this study are (1) the performance of the binary MLP with 32 hidden units was inferior to that of the real-valued MLP with 16 hidden units; however, this difference was not statistically significant (p >.05); (2) the trained binary MLP successfully swung the pendulum swiftly into an inverted position and maintained its stability after inversion, as the real-valued MLP had done; and (3) the memory size required to record the binary MLP with 32 hidden units is 3.1% or 6.2% of the memory size required to record the real-valued MLP with 16 hidden units.

Defect detection plays a pivotal role in quality control for fabrics. In order to enhance the accuracy and efficiency of fabric defect detection, we have proposed the PRC-Light YOLO model for fabric defect detection and have established a detection system. Firstly, we add new convolution operators in Backbone for the YOLOv7 and integrate them with Extended-Efficient Layer Aggregation Network. This combination constructs a new feature extraction module that not only reduces the computational effort of the network model, but also extracts spatial features effectively. Secondly, we employ multi-branch dilated convolutions feature pyramid, and introduce lightweight upsampling operators to improve performance of the feature fusion network. This module achieves an expanded receptive field by generating real-time adaptive convolution kernels, allowing for the collection of crucial information from regions with a richer contextual context. To further minimize the computation during network model training, we adopt the HardSwish activation function. Finally, we apply the Wise-IOU v3 bounding box loss function as a dynamic non-monotonic focusing mechanism, which decreases adverse gradients from low-quality instances. We conduct data augmentation on real fabric dataset to raise the generalization capability of the PRC-Light YOLO model. Compared with the YOLOv7 model, numerous simulation experiments show that our proposed methods reduce the model’s parameters and computation by 18.03% and 20.53%, respectively. Simultaneously, there has been a 7.6% improvement in mAP.

The delineation of the clinical target volumes (CTVs) for radiation therapy is time-consuming, requires intensive training and shows high inter-observer variability. Supervised deep learning methods depend heavily on consistent training data, thus state-of-the-art research focuses on making CTV labels more homogeneous and strictly bounding them to current standards. International consensus expert guidelines standardize CTV delineation by conditioning the extension of the clinical target volume on surrounding anatomical structures. Training strategies that directly following the construction rules given in the expert guidelines or the possibility to quantify the conformance of manually drawn contours with the guidelines are still missing. 71 anatomical structures that are relevant for CTV delineation in head and neck cancer patients according to the expert guidelines were segmented on 104 CT scans to assess the possibility to automate their segmentation by state-of-the-art deep learning methods. nnU-Net models were trained to automatically segment those 71 structures on planning CT scans. We report the DICE, HD and sDICE for 71 + 5 anatomical structures for most of which no previous segmentation accuracies have been reported. For those structures, for that prediction values have been reported, our segmentation accuracy matched or exceeded the reported values. The predictions from our models were always better than those predicted by the TotalSegmentator. The sDICE with 2mm margin was larger than 80% for almost all structures. Individual structures with decreased segmentation accuracy are analyzed and discussed with respect to their impact on the CTV delineation following the expert guidelines. No deviation is expected to affect the rule-based automation of the CTV delineation.

In this study, we present a federated learning approach that combines a multi-branch network and the Oort client selection algorithm to improve the performance of federated learning systems. This method successfully addresses the significant issue of non-iid data, a challenge not adequately tackled by the commonly used MFedAvg method. Additionally, one of the key innovations of this research is the introduction of uniformity, a metric that quantifies the disparity in training time amongst participants in a federated learning setup. This novel concept not only aids in identifying stragglers but also provides valuable insights into assessing the fairness and efficiency of the system. Experimental results underscore the merits of the integrated multi-branch network with the Oort client selection algorithm and highlight the crucial role of uniformity in designing and evaluating federated learning systems.

Intelligent Tutoring Systems (ITS) are increasingly popular for online learning. These systems use adaptive algorithms to recommend relevant content based on students' profiles. However, instructors need to periodically assess students' performance to ensure learning outcomes and adjust strategies accordingly. Our objective is to predict students' progress in advance, enabling teachers to make quicker decisions and facilitating the iterative process of adaptive algorithms. For this study, we collected a dataset from ALIN, an online learning platform, consisting of over 5,000 students' learning records and test results. Using this dataset, we conducted experiments employing various machine learning algorithms. The results indicate that learning behavior contributes to improving forecast performance, while students' progress strongly correlates with their previous test results. Additionally, we discovered that students' progress can be indirectly predicted by forecasting their scores. Furthermore, by breaking down overall scores into several distinct components and predicting individual scores for each component, the accuracy of the forecasts can be improved.

The paradigm of leveraging code as a dataset has recently gained traction, offering innovative solutions in domains such as automated commit message generation, pull request description automation, and program repair mechanisms. Consider the challenge in generating commit messages: traditional methods treat source code as a mere token sequence, applying neural machine translation models. This approach, however, overlooks the critical syntactic structures inherent in programming languages, which could offer deeper insights and improved accuracy. Building upon prior research, specifically the Code2Seq framework, which utilized Abstract Syntax Tree (AST) structural data for source code representation to automate method name generation, this paper extends and refines this concept. We introduce "CSR", a novel methodology adapted to represent code edits effectively. This paper investigates the impact of employing syntactic structure, focusing on the classification of code edits. Drawing inspiration from Code2Seq, "CSR" utilizes AST's structural properties, particularly the paths connecting leaf nodes, to enhance the task of code edit classification. This approach is rigorously evaluated on two distinct datasets, comprising fine-grained syntactic edits. Our comprehensive experiments reveal that incorporating syntactic structures does not significantly outperform simpler methodologies. While methods like Code2Seq and our proposed "CSR" show potential, our findings highlight that there is considerable scope for improvement and refinement before such techniques can be universally applied for learning representations of code edits. We anticipate that our findings will spark further research in this field, paving the way for more effective use of syntactic structures in code representation.

Machine learning is a fundamental aspect of artificial intelligence that involves the development of algorithms and models that allow computers to learn and make predictions or decisions without explicit programming. With the development of neural networks, back-propagation algorithms and deep learning, machine learning has made breakthroughs in the fields of image recognition, natural language processing and handwriting recognition using machine learning techniques. The advent of deep learning has revolutionized the field of handwriting recognition, using convolutional neural networks, recurrent neural networks, and sequence-to-sequence models to provide solutions that go beyond machine learning methods and significantly improve the accuracy and robustness of handwriting recognition systems. But challenges remain, including the need for large labelled datasets, computational resources and addressing potential biases. As research in deep learning techniques continues to drive handwriting recognition closer towards realisability, machine learning approaches remain at the forefront.

The energy-intensive sector of the cement industry needs to find technologically advanced methods to produce cement with as little energy as possible, without polluting the environment and without sacrificing the quality (fineness) of the cement. In addition, the stress on the machines changes the behavior of the data. By combining the k-means method, two (2) feature selection methods, nine (9) machine learning methods, and the differential evolution (DE) method, a dynamic system is created where the user determines the key performance indicators (KPIs) each time. The user has the potential to control the distribution of the data, the input of machine learning models, and the ability to switch the system to the suggested values without damage. The selection of the best-performing models is based on the normalized root mean squared error (NRMSE). The DE method optimizes the dynamic objective function. The optimal values of the manipulated variables are evaluated compared to the value of the objective function from the actual values of the variables. In all experiments, the recommendations are acceptable. The final results show an improvement in the operation of both the cement mill and kiln compared to the current operating condition.

Activities of daily living (ADLs) are fundamental routine tasks that the majority of physically and mentally healthy people can independently execute. In this paper, we present a Semantic Framework for detecting problems in ADLs execution, monitored through Smart Home sensors. In the context of this work, we conducted a pilot study, gathering raw data from various sensors and devices installed in a smart home environment. The proposed Framework combines multiple Semantic Web technologies (i.e. ontology, RDF, triplestore) to handle and transform these raw data into meaningful representations forming a Knowledge Graph. Subsequently, SPARQL queries are used to define and construct explicit rules to detect problematic behaviours in ADLs execution, a procedure that leads to generating new implicit knowledge. Finally, all available results are visualized in a clinician dashboard. The proposed framework can monitor the deterioration of ADLs performance for people in the spectrum of dementia, by offering a comprehensive way for clinicians to describe problematic behaviors in the every-day life of an individual.

Contemporary advanced models for aspect-targeted opinion word extraction (ATOWE), which predominantly utilize BERT-based encoders at a word level, have shown limited advancements when integrated with graph convolutional networks (GCNs) for syntactic tree assimilation. Recognizing the prowess of BERT subwords in encapsulating rare or context-poor words, this study pivots from syntactic trees to BERT subwords, omitting GCNs from the structural framework. Our approach, named Aspect-Enhanced Wordpiece Extraction Model (AEWEM), focuses on refining aspect representation during encoding. We propose an input format of paired sentence-aspect, diverging from traditional single-sentence inputs. AEWEM demonstrates superior performance on benchmark datasets, establishing a robust foundation for future explorations in this domain.

Squash is a sport where referee decisions are essential to the game. However, referee decisions in squash are very subjective in nature. Disputes, both from the players and the audience, regularly happen because the referee made a controversial call. In this study ,we proposed to automate the referee decision process through machine learning. We trained neural networks to predict squash referee decisions using data from 400 referee decisions acquired through extensive video footage reviewing and labeling. Six positional values were extracted, including the attacking player’s position, the retreating player’s position, the ball’s position in the frame, the ball’s projected first bounce, the ball’s projected second bounce, and the attacking player’s racket head position. We calculated nine additional distance values, such as the distance between players and the distance from the attacking player's racket head to the ball's path. Models were trained on Wolfram Mathematica and Python using these values. The best Wolfram Mathematica model achieved an 86% ± 3.03% accuracy, while the best Python model performed a 0.852 ± 0.051 accuracy (85.2% ± 5.1%). These accuracies surpass 85%, demonstrating near-human performances. Our model has great potential for improvement as it's currently trained with limited data (400 referee decisions) and lacks crucial data points such as time and speed. The performance of our model is almost surely going to improve significantly with a larger training data set. Unlike human referees, machine learning models follow a consistent standard, have unlimited attention spans, and make decisions instantly. If the accuracy is improved in the future, the model can potentially serve as an extra refereeing official for both professional and amateur squash matches. Both the analysis of referee decisions in squash and proposal to automate the process using machine learning is unique to this study.

The World Health Organization (WHO) has released reports indicating that heart disorders hold the unfortunate distinction of being the primary cause of death worldwide. Shockingly, an astonishing estimated 17.9 million lives are claimed by heart diseases annually, accounting for an alarming 31% of all global deaths. With all the flaws in the clinical situation, it is frequently challenging to assess the severity of cardiac disease and forecast its course of progression due to heterogeneity and complex interplay of factors. Therefore, early heart disease detection is essential for effective therapy. To address these challenges, Machine Learning (ML) boosting algorithms play a pivotal role as the main components of predictive analytics required to do this. The main objective of this study is to develop a comprehensive comparative framework to predict heart diseases using state-of-the-art machine learning with boosting techniques such as Decision Tree, Random Forest, Gradient Boosting, Catboost, XGboost, Light GBM, and Adaboost. To evaluate the performance of the models, a large heart disorders dataset is used from the UCI machine learning repository, comprised of 26 feature-based numerical and categorical attributes with 8763 samples from all over the globe. The Experimental results reveal that AdaBoost attained the highest accuracy of 95% and outperforms other algorithms concerning various performance measures like precision= 0.98, recall= 0.95, specificity= 0.95, f1-score=0.01, Negative predicted value= 0.83, False positive rate= 0.04, False negative rate= 0.04 and False Development rate= 0.01.

Artificial Intelligence (AI) in healthcare has ushered in a revolutionary era by transforming the way we provide care and reduce adverse consequences. This paper explores the field of AI applications in healthcare, including side effect management, streamlined administrative procedures, and customized treatment programs.

The art of representing source code is pivotal in numerous programming analysis applications. Recent strides in neural networks have marked notable successes in this realm. However, the peculiar structural characteristics inherent in programming languages have not been fully exploited in existing models. While neural models based on abstract syntax trees (ASTs) adeptly manage the tree-like nature of source codes, they fall short in discerning the diverse substructural nuances within programs. This paper introduces the Dynamic Syntax Tree Model (DSTM), an innovative approach that fuses various neural network modules into tree architectures tailored to the specific AST of the input. Distinct from preceding tree-based neural models, DSTM adeptly discerns the semantic variances across different AST substructures. We validate DSTM through rigorous testing in program classification and code clone detection, outperforming contemporary methods and demonstrating the benefits of harnessing intricate source code structures.

To support pathologists in breast tumor diagnosis, deep learning plays a crucial role in the development of histological whole slide image (WSI) classification methods. However, automatic classification is challenging due to the high-resolution data and the scarcity of representative training data. To tackle these limitations, we propose a deep learning-based breast tumor gigapixel histological image multi-classifier integrated with a high-resolution data augmentation model to process the entire slide by exploring its local and global information and generating its different synthetic versions. The key idea is to perform classification and augmentation in feature latent space, reducing the computational cost while preserving the class label of the input. We adopt a deep learning-based multi-classification method and evaluate the contribution given by a Conditional Generative Adversarial Network-based data augmentation model on the classifier’s performance for three tumor classes in the BRIGHT Challenge dataset. The proposed method has allowed us to achieve an average F1 equal to 69.5, considering only the WSI dataset of the Challenge. The results are comparable to those obtained by the Challenge winning method (71.6), also trained on the annotated tumor region dataset of the Challenge.

The article focuses on breast cancer, mammography, and artificial intelligence. First, breast cancer is a widespread health problem that affects millions of people worldwide, and mammography is a widely adopted screening method. Then it introduced the advantages of mammography after AI participation, and introduced the importance of early detection of breast cancer and the application of artificial intelligence in breast cancer detection and treatment. From these aspects extend to the entire medical field of artificial intelligence issues. Several times throughout the paper, ethical issues that could arise from applying AI to healthcare are highlighted. At the end of the article, the paper describes the continued development of AI-based mammography analysis over the next period of time.

Sleep disorder is a disease that can be categorized as both an emotional and physical problem. It imposes several difficulties and problems, such as distress during the day, sleep-wake disorders, anxiety, and several other problems. Hence, the main objective of this research is to utilize the strong capabilities of machine learning in the prediction of sleep disorders. In specific, this research aims to meet three main objectives. These objectives are to identify the best regression model, the best classification model, and the best learning strategy that highly suits sleep disorder datasets. Considering two related datasets and several evaluation metrics that are related to the tasks of regression and classification, the results revealed the superiority of the MultilayerPerceptron, SMOreg, and KStar regression models compared with the other twenty-three regression models. Also, IBK, RandomForest, and RandomizableFilteredClassifier showed superior performance compared with other classification models that belong to several learning strategies. Finally, the Function learning strategy showed the best predictive performance among the six considered strategies in both datasets and with respect to the most evaluation metrics.

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.

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.

The Transformer architecture, while adept at capturing context through self-attention, falls short in encapsulating complex syntactic structures effectively. Addressing this gap, we introduce the Linguistic Structure through Graphical Interpretation with BERT (LSGIB) approach in Machine Translation (MT) frameworks. Combining the strengths of Graph Attention Network (GAT) and BERT, LSGIB intricately captures syntactic dependencies as explicit knowledge from the source language. This enhances the source language representation and aids in more accurate target language generation. Our empirical analysis leverages gold-standard syntax-annotated sentences and employs a Quality Estimation (QE) model. This approach enables us to assess translation improvements in terms of syntactic accuracy, extending beyond traditional BLEU score metrics. The LSGIB model demonstrates superior translation quality across diverse MT tasks, maintaining robust BLEU scores. Our study delves into the optimal sentence lengths benefiting from LSGIB and identifies which syntactic dependencies are more precisely captured. We observe that GAT's ability to learn specific dependency relations directly influences the translation quality of sentences with those relations. Additionally, we discover that incorporating syntactic structure into BERT's intermediate and lower layers offers a novel approach to modeling linguistic structure in source sentences.

Public health is an important aspect of community activities, making research on health necessary because it is a crucial field in maintaining and improving the quality of life in society as a whole. Research in public health allows for a deeper understanding of the health problems faced by a population, including disease prevalence, risk factors and other determinants of health. This work aims to explore the potential of hospital patient data analysis as a valuable tool for understanding community implications and deriving insights for effective community health interventions. The study recognises the significance of harnessing the vast amount of data generated within hospital settings to inform population-level health strategies. The methodology employed in this study involves the collection and analysis of deidentified patient data from a representative sample of a hospital in Indonesia. Various data analysis techniques, such as statistical modelling, data mining and machine learning algorithms, are utilised to identify patterns, trends and associations within the data. A program written in Python is used to analyse patient data in a hospital for five years from 2018 to 2022. These findings are then interpreted within the context of public health implications, considering factors such as disease prevalence, socioeconomic determinants and healthcare utilisation patterns. The results of the data analysis provide valuable insights into the public health implications of hospital patient data. The research also covers prediction for the patient data to the hospital based on disease, age and geographical residence. The research prediction shows that in 2023, the number of patients was not considerably affected, but in in March to April 2024, the number significantly increased to 10,000 patients. These recommendations encompass targeted prevention strategies, improved healthcare delivery models and community engagement initiatives. The research emphasises the importance of collaboration between healthcare providers, policymakers and community stakeholders in implementing and evaluating these interventions.

A new approach for predicting the geometrical characteristics of the mixing behavior of an inclined dense jet for angles ranging from 15° to 85° is proposed in this study. This approach is called group method of data handling (GMDH) which is based on the artificial neural network (ANN) technique. The proposed model was trained and tested using existing experimental data reported in literature. The model was then evaluated using statistical indices as well as compared with analytical models from previous studies. The results of the coefficient of determination (R2) indicate a high accuracy of the proposed model with values of 0.9719 and 0.9513 for training and testing for the dimensionless of the distance from the nozzle to the return point xr/D, and 0.9454 and 0.9565 for training and testing for the dimensionless of the terminal rise height yt/D . Moreover, four previous analytical models were used to evaluate the GMDH model. The results showed the superiority of the proposed model in predicting the geometrical characteristics of the inclined dense jet for all tested angles. Finally, the standard error of estimate (SEE) was applied to demonstrate which model performed the best in terms of getting closer to the actual data. The results illustrate that all fitting lines of the GMDH model performed very well for all geometrical parameter predictions and was the best model with approximately 10% error, which was the lowest value of error among the models. Therefore, this study confirms that the GMDH model can be used to predict the geometrical properties of the inclined negatively buoyant jet with high performance and accuracy.

Using metaheuristics, such as the Evolutionary Algorithm (EA), within control structures is a realistic approach for certain optimal control problems. Their role is often predicting the optimal control values over a prediction horizon using a process model (PM). The big computational effort sometimes causes problems. Our work addresses a new issue: whether a machine learning (ML) algorithm could “learn” the optimal behaviour of the couple (EA, PM). A positive answer is given by proposing datasets apprehending this couple's optimal behaviour and appropriate ML models. Following a design procedure, a number of closed-loop simulations will provide the sequences of optimal control and state values, which are collected and aggregated in a data structure. For each sampling period, datasets are extracted from the aggregated data. The ML algorithm experiencing these datasets will produce a set of regression functions. Replacing the EA predictor with the ML model, new simulations are carried out, proving that the state evolution is almost identical. The execution time decreases drastically because the PM’s numerical integrations are totally avoided. The performance index equals the best-known value. In different case studies, the ML models succeeded in capturing the optimal behaviour of the couple (EA, PM) and yielded efficient controllers.

In the evolving landscape of Portfolio Management (PM), the fusion of advanced machine 1 learning techniques with traditional financial methodologies has opened new avenues for innovation. 2 Our study introduces a cutting-edge model combining Deep Reinforcement Learning (DRL) with 3 a Non-stationary Transformer architecture. This model is specifically designed to decode complex 4 patterns in financial time series data, enhancing portfolio management strategies with deeper insights 5 and robustness. It effectively tackles the challenges of data heterogeneity and market uncertainty, 6 key obstacles in PM. Our approach integrates key macroeconomic indicators and targeted news 7 sentiment analysis into its framework, capturing a comprehensive picture of market dynamics. This 8 amalgamation of varied data types addresses the multifaceted nature of financial markets, enhancing 9 the model’s ability to navigate through the complexities of asset management. Rigorous testing 10 demonstrates the model’s efficacy, highlighting the benefits of blending diverse data sources and 11 sophisticated algorithmic approaches in mastering the nuances of PM.12

Few-shot fine-grained image classification (FSFGIC) methods refer to machine learning methods which aim to classify images (e.g., bird species, flowers, and airplanes) belonging to subordinate object categories of the same entry-level category with only a few samples. It is worth to note that feature representation learning is used not only to represent training samples, but also to construct classifiers for performing various FSFGIC tasks. In this paper, starting from the definition of FSFGIC, a taxonomy of feature representation learning for FSFGIC is proposed. According to this taxonomy, we discuss key issues on FSFGIC (including data augmentation, local or/and global deep feature representation learning, class representation learning, and task specific feature representation learning). The existing popular datasets and evaluation standards are introduced. Furthermore, a novel classification performance evaluation mechanism is designed with a 0.95 confidence interval for judging whether the classification accuracy obtained by a certain specified method is good or bad. Moreover, current challenges and future trends of feature representation learning on FSFGIC are elaborated.

This work develops a new method for vector data augmentation. The proposed method applies the Principal Component Analysis (PCA) and determines the Eigenvectors of a set of training vectors for a Machine Learning (ML) method and uses them to generate the distilled vectors. The training and the PCA distilled vectors have the same dimension. The user chooses the number of vectors to be distilled and augmented to the set of training vectors. A statistical approach determines the lowest number of vectors to be distilled such that when augmented to the original vectors the extended set trains a ML classifier to achieve a required accuracy. Hence, the novelty of this study is, the distillation of vectors with the PCA method and their use to augment the original set of vectors. The advantage that comes from the novelty is that it increases the statistics of ML classifiers. To validate the advantage, we conducted experiments with four public databases and applied four classifiers: Neural Network, Logistic Regression, Support Vector Machine with linear and polynomial kernels. For the purpose of augmentation, we conduct several distillations including nested (double distillation). The latter notion means that new vectors were distilled from already distilled vectors. We train the classifiers with three sets of vectors: original vectors; original augmented with vectors distilled by PCA; original augmented with distilled and double distilled, by PCA, vectors. The experimental results are presented in the paper, and they confirm the advantage that the PCA distilled vectors increase the classification statistics of ML methods, if the distilled vectors augment the original training vectors.

Post-training quantization is pivotal in deploying convolutional neural networks for mobile applications. Block-wise reconstruction with adaptive rounding, as employed in prior works like BrecQ and Qdrop, facilitates acceptable 4-bit quantization accuracy. However, adaptive rounding is time-intensive, and its constraint on weight optimization space curtails the potential for quantization performance. The optimality of block-wise reconstruction hinges on the quantization status of subsequent network blocks. In this investigation, we delve into the theoretical underpinnings of the limitations inherent in adaptive rounding and block-wise reconstruction. Our exploration leads to the development of a post-training quantization methodology, designated as AE-Qdrop. This algorithm operates in two distinct phases: block-wise reconstruction and global fine-tuning. The block-wise reconstruction phase introduces a progressive optimization strategy, superseding adaptive rounding, which not only augments quantization precision but also significantly improves quantization efficiency. To mitigate the risk of overfitting, we introduce a random weighted quantized activation mechanism. During the global fine-tuning phase, we account for interdependencies among quantized network blocks. The weight of each network block will be corrected with logit matching and feature matching. Extensive experiments validate that AE-Qdrop achieves high-precision and efficient quantization. For instance, in the case of 2-bit MobileNetV2, AE-Qdrop outperforms Qdrop by achieving a 6.26% enhancement in quantization accuracy and quintupling the quantization efficiency.

Audio classification using deep learning models, essential for applications like voice assistants and music analysis, faces challenges when deployed on edge devices due to their limited computational resources and memory. Achieving a balance between performance, efficiency, and accuracy is a significant obstacle in optimizing these models for such constrained environments. In this investigation, we evaluate diverse deep learning architectures, including Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM), for audio classification tasks on the ESC 50, UrbanSound8k, and Audio Set datasets. Our empirical findings indicate that Mel Spectrograms outperform raw audio data, attributing this enhancement to their synergistic alignment with advanced image classification algorithms and their congruence with human auditory perception. To address the constraints of model size, we apply model compression techniques, notably magnitude pruning, Taylor Pruning, and 8-bit quantization. The research demonstrates that a hybrid pruned model achieves a commendable accuracy rate of 89 percent, which, although marginally lower than the 92 percent accuracy of the uncompressed CNN, strikingly illustrates an equilibrium between efficiency and performance. Subsequently, we deploy the optimized model on Raspberry Pi 4 and NVIDIA Jetson Nano platforms for audio classification tasks. These findings highlight the significant potential of model compression strategies in enabling effective deep learning applications on resource-limited devices, with minimal compromise on accuracy.

The purpose of this study is to ​​explore the use of a custom Transformer model in brain-computer interfaces (BCIs) that translate the neural activity present when an individual with limited verbal and fine-motor skills attempts to handwrite. As found in previous studies, Transformers have performed better than recurrent neural networks (RNNs) in translation tasks which are similar to its usage here in decoding neural signals into intended handwritten text. Due to the known benefits of a Transformer, the hypothesis was that the Transformer would show promise in the context of a BCI through the recorded metrics. The neural signals of a tetraplegic individual, when they attempted to handwrite, were provided by existing research. Four trials were conducted using data with or without augmentation which the model used when training to separately minimize training loss and validation loss. When comparing the results of a BCI with the implementation of a Transformer model with the original RNN BCI (the original data source), the Transformer model performed less favorably across all four trials. Although the results do not indicate that the Transformer model currently outperforms an RNN BCI, it is important to note that further testing of the model's capabilities (such as training it with a larger and more preferable dataset and/or for longer, comparing training times between the RNN and Transformer, and/or seeing how the Transformer is improved with an offline autocorrect feature) is necessary before determining whether Transformers can enhance communication in this manner.

The evolution of sentence representation learning, especially with parse tree encoders, has shown remarkable progress. Traditional approaches predominantly rely on recursive encoding of tree structures, which impedes parallel processing capabilities. Additionally, these methods often overlook the significance of dependency tree arc labels. To overcome these limitations, we introduce the Syntax-Enhanced Transformer (SET), incorporating a novel dual-attention mechanism that integrates relation-focused attention alongside traditional self-attention. This design effectively encodes both dependency and spatial positional relationships within sentence dependency trees. Our approach innovatively incorporates syntactic information into the Transformer framework without compromising its inherent parallelizability. The SET demonstrates superior or comparable performance to contemporary methods across various sentence representation tasks, significantly enhancing computational efficiency.

In recent years, Edge Intelligence (EI) has emerged, combining edge computing with AI, specifically deep learning, to run AI algorithms directly on edge devices. In practical applications, EI faces challenges related to computational power, power consumption, size, and cost, with the primary challenge being the trade-off between computational power and power consumption. This has rendered traditional computing platforms unsustainable, making heterogeneous parallel computing platforms a crucial pathway for implementing EI. In our research, we leveraged the Xilinx Zynq 7000 heterogeneous computing platform, employed High-Level Synthesis (HLS) for design, and implemented two different accelerators for LeNet-5 using loop unrolling and pipelining optimization techniques. The experimental results show that when running at a clock speed of 100 MHz, the PIPELINE accelerator, compared to the UNROLL accelerator, experiences an 8.09% increase in power consumption but speeds up by 14.972 times, making the PIPELINE accelerator superior in performance. Compared to the CPU, the PIPELINE accelerator reduces power consumption by 91.37% and speeds up by 70.387 times, while compared to the GPU, it reduces power consumption by 93.35%.This study provides two different optimization schemes for edge intelligence applications through design and experimentation, and demonstrates the impact of different quantization methods on FPGA resource consumption. These experimental results can provide a reference for practical applications, thereby providing a reference hardware acceleration scheme for edge intelligence applications.

In the present article, we study the possibilities of machine learning for estimation of seeing at the Maidanak Astronomical Observatory (38∘40′24′′N, 66∘53′47′′E) using only Era-5 re-analysis data. Seeing is usually associated with the integral of the turbulence strength Cn2(z) over the height z. Based on the seeing measurements accumulated over 13 years we created ensemble models of multi-layer neural networks under the machine learning framework, including training and validation. For the first time in the world, we have simulated optical turbulence (seeing variations) during night-time with deep neural networks trained on a 13-year database of astronomical seeing. A set of neural networks for simulations of night-time seeing variations have been obtained. For these neural networks the linear correlation coefficient ranges from 0.48 to 0.68. We show that modeled seeing with neural networks is well described through meteorological parameters, which include wind speed components, air temperature, humidity and turbulent surface stresses. One of the fundamental new results is that the structure of small-scale (optical) turbulence over the Maidanak Astronomical Observatory does not depend or depends negligibly on the vortex component of atmospheric flows.

Hyperspectral (HS) imaging plays a pivotal role in various fields, including medical diagnostics, where precise human vein detection is crucial. Hyperspectral image data are very large and can cause computational complexities. Dimensionality reduction techniques are often employed to streamline HS image data processing. This paper investigates the effectiveness of three dimensionality reduction techniques, namely: Principal Component Analysis (PCA), Folded PCA (FPCA), and Ward’s Linkage Strategy using Mutual Information (WaLuMI) for vein detection using HS images. A HS image dataset, encompassing left and right-hand images captured from 100 subjects with varying skin tones was created and annotated using anatomical data to represent vein and non-vein areas within the images. To generate experimental results, the HS image dataset was divided into train and test datasets. Optimum performing parameters for each of the dimensionality reduction techniques in conjunction with The Support Vector Machine binary classification were determined using the Training dataset. The performance of the three dimensionality reduction based vein detection methods was then assessed and compared using the test image dataset. Results show that the FPCA-based method outperforms the other two methods in terms of accuracy. For visualization purposes, the classification prediction image for each technique is post-processed using morphological operators, and results show the significant potential of HS imaging in vein detection.

Despite being considered as natural components of many ecosystems, forest fires pose significant threats to the environment and human health. In order to ensure public safety and effective fire suppression planning, it is necessary to develop reliable prediction models to mitigate forest fire danger. These models should account for specific environmental conditions. The advent of big data in recent years has opened new avenues for improving forest fire predictions. Machine learning techniques, surpassing traditional forecasting methods, have shown significant promise in this area. By applying the Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology to a real-world dataset, this paper explores the application of machine learning approaches to understand forest fire patterns and predict fire danger. We consider six distinct time stages and incorporate feature selection to refine our predictions. It is important to note that forest fire behavior models are not universally effective due to geographical variations in data. Nevertheless, advanced decision-making techniques are vital in forest fire management. Our research presents a systematic exploration of the topic, comparing various machine learning models, thereby providing a comprehensive baseline for future investigations in this crucial environmental arena.

Remediation of IT issues encoded into domain-specific or user-defined alerts occurring in cloud environments and customer ecosystems in a vast majority of cases suffers from accurate recommendations which could be timely supplied for recovery of performance degradations. That, of course, is hard to realize through furnishing those abnormality definitions with an appropriate expert knowledge which varies from one environment to another. At the same time, in a large proportion of support cases, the reported problems under Global Support Services (GSS) or Site Reliability Engineering (SRE) treatment ultimately go down to the product teams making them waste costly development hours on investigating self-monitoring metrics of our solutions. Therefore, the mean-time-to-resolution (MTTR) rates of problems/alerts are significantly impacted from lack of a systematic approach towards adopting AI Ops. That would imply building, maintaining, and continuously improving/annotating a data store of insights that ML models are trained and generalized across the whole customer base and corporate cloud services. Our ongoing study is in line with such a vision and validates an approach that learns the alert resolution patterns in such a global setting and explains them using interpretable AI methodologies. The knowledge store of causative rules is then applied in predicting potential sources of the application degradation reflected in an active alert instance. In this communication, we share our experiences with a prototype solution and up to date analysis demonstrating how root conditions are discovered with a high accuracy for a specific type of problem. It is validated against the historical data of resolutions performed by heavy manual development efforts. We also offer a Dempster-Shafer theory-based rule verification framework for experts as a what-if analysis tool to test their hypotheses about underlying environment.

With the development of the Internet, The content people share contains types of text, images, and videos, and utilizing these multimodal data for sentiment analysis has become an important area of research. Multimodal sentiment analysis aims to understand and perceive emotions or sentiments in different types of data. Currently, the realm of multimodal sentiment analysis faces various challenges, with a major emphasis on addressing two key issues: 1) Inefficiency when modeling the intra-modality and inter-modality dynamics and 2) Inability to effectively fuse multimodal features. In this paper, we proposed the CCDA(Cross-Correlation in Dual-Attention) model, a novel method to explore dynamics between different modalities and fuse multimodal features efficiently. We capture dynamics at intra- and inter-modal levels by using two types of attention mechanisms simultaneously. Meanwhile, the cross-correlation loss is introduced to capture the correlation between attention mechanisms. Moreover, the relevant coefficient is proposed to integrate multimodal features effectively. Extensive experiments were conducted on three publicly available datasets, CMU-MOSI, CMU-MOSEI, and CH-SIMS. The experimental results fully confirm the effectiveness of our proposed method, and compared with the current optimal method (SOTA), our model shows obvious advantages in most of the key metrics, proving its better performance in multimodal sentiment analysis.

Calligraphy works have high artistic value, but there is a rampant problem of forgery. Indeed, authentication of traditional calligraphy heavily relies on calligraphers' subjective judgment. Therefore, spurred by the recent development of neural networks, this paper proposes a method for authenticating calligraphy works based on an improved EfficientNet network. Specifically, the developed method utilizes the character box algorithm and the centroid algorithm to extract individual calligraphy characters, which are then augmented and used as the training set for the model. The training process employs CBAM and Self-Attention modules to enhance the attention mechanism of the EfficientNet network. The model is tested on authentic works, imitated works, and works from other calligraphers and is compared with other networks. The experimental results demonstrate that the proposed method effectively achieves the authentication of calligraphy works, and the improved CBAM-EfficientNet network and SA-EfficientNet network achieve better authentication performance.

This paper introduces a novel graph polynomial approach for differentiating tree structures in dependency grammar. Utilizing this polynomial representation, we develop a metric to assess the similarity in syntax. This approach offers a detailed and inclusive analysis of the dependency structures and relationships in sentence construction. We employ this polynomial method to examine sentence structures across various languages in the Parallel Universal Dependencies treebanks. Our analysis includes comparing the syntax of original sentences and their translated counterparts in diverse languages, alongside a comprehensive study of syntactic typologies within these treebanks. Additionally, we explore the application of our methodology in evaluating the syntactic diversity within language corpora.

Video lectures are becoming more popular and in demand as online classroom teaching is becoming more prevalent. Massive Open Online Courses (MOOCs), such as NPTEL, have been creating high-quality educational content that is freely accessible to students online. A large number of colleges across the country are now using NPTEL videos in their classrooms. So more video lectures are being recorded, maintained, and uploaded. These videos generally contain information about that video before the lecture begins. We generally observe that these educational videos have metadata containing five to six attributes: Institute Name, Publisher Name, Department Name, Professor Name, Subject Name, and Topic Name. It would be easy to maintain these videos if we could organize them according to their categories. The indexing of these videos based on this information is beneficial for students all around the world to efficiently utilise these videos. In this project, we are trying to get the metadata information mentioned above from the video lectures.

The process of Targeted Opinion Word Extraction (TOWE), a critical component of aspect-based sentiment analysis (ABSA), revolves around identifying opinionated words linked to specific aspect-terms within sentences. Existing deep learning approaches, while effective, often overlook the syntactic structure of sentences, a factor that previous studies have identified as beneficial for TOWE. In this study, we introduce the Syntactic-Enhanced Deep Learning Model (SEDLM) that integrates syntactic structures into deep learning frameworks for TOWE. Our approach leverages syntax-driven opinion potential scores and syntactic inter-word connections, enhancing model performance. Additionally, we introduce an innovative regularization strategy aimed at distinguishing word representations in TOWE tasks. Our comprehensive analysis reveals that SEDLM sets new benchmarks in performance across multiple standard datasets.

It is essential to deal with data scattered among heterogeneous information sources, which can be structured, semi-structured, or unstructured, to give consumers a cohesive perspective of the data. Information gathering is challenging as a result, and one of the major causes of this is that data sources are developed to support specific applications. A method to simplify this procedure is to use ontology representation as an intermediate step. An ontology represents a knowledge structure that reasonably reflects the real world's complexity and is ideally used in many industries today. This research aims to enhance the functionality of a global ontology system by implementing and integrating a natural language query system. Leveraging NLP-based approaches, we constructed SPARQL queries for our ontologies to facilitate natural language translation into a well-structured query format. A dedicated query component was developed to transform natural language questions into SPARQL queries, with the selection of the optimal query graph to generate the final queries. In the subsequent phase, we explored the integration of distinct sentence encoders to improve latent sentence representations during query construction. The generated SPARQL queries were executed on the global ontology output and then translated into source-specific questions. This innovative approach enables unified access to heterogeneous data sources through a user-friendly natural language querying interface.

Recent advancements in neural language models, particularly attention-based architectures like the Transformer, have substantially surpassed traditional methods in various natural language processing tasks. These models adeptly generate nuanced token representations by considering contextual relations within a sequence. However, augmenting these models with explicit syntactic knowledge, such as part of speech tags, has been found to remarkably bolster their effectiveness, especially under constrained data scenarios. This study introduces the Linguistic Enhanced Transformer (LET), which integrates multiple syntactic features, showing a notable increase in translation accuracy, evidenced by an improvement of up to 1.99 BLEU points on subsets of the WMT '14 English-German dataset. Furthermore, this paper demonstrates that enriching BERT models with syntax-aware embeddings enhances their performance on several GLUE benchmark tasks.

Conventional text style transfer (TST) methodologies primarily utilize style classifiers to segregate the content and stylistic elements of text for effective style transformation. Despite the pivotal role of these classifiers, their influence on TST techniques remains largely unexplored. This study embarks on a detailed exploration of the limitations inherent in style classifiers within current TST frameworks. We reveal that these classifiers often inadequately comprehend sentence syntax, leading to diminished performance in TST models. In response, we introduce the Syntax-Enhanced Style Transfer (SEST) model, a groundbreaking approach incorporating a syntax-sensitive style classifier. This classifier ensures that the extracted style representations robustly encapsulate syntax nuances, enhancing TST effectiveness. Rigorous evaluations across diverse TST benchmarks demonstrate that SEST significantly surpasses contemporary models in performance. Additionally, our case studies highlight SEST's proficiency in producing syntactically coherent sentences that aptly retain original content.

With the development of sensor and communication technologies, Internet of Things (IoT) subsystem is gradually becoming a crucial part in vehicles. It can effectively enhance functionalities of vehicles. However, new attack types are also emerged. For example, a driver with the smart key in the pocket can push the start button to start a car when. At the same time, security issues in the push-to-start scenario are alerted, such as smart key forgery. In this paper, we propose a vehicle Passive Entry Passive Start (PEPS) system that uses deep learning algorithms to recognize the driver by the Electrocardiogram (ECG) signals measured by his or her smart watch. ECG signals are used for verification. Smart watches, as a new smart key of PEPS system, can replace traditional smart keys to improve security. Experiment results show that Long Short-Term Memory (LSTM) models have achieved the best accuracy score for identity recognition (91%) when single ECG cycle is used. However, it takes at least 30 minutes for training. The training time of Auto Encoder successfully reduces to 5 minutes. When 15 continuous ECG cycles are used, it can achieve 100% identity accuracy.

Machine learning approaches are currently used to understand or model complex physical systems. In general, a substantial number of samples must be collected to create a model with reliable results. However, collecting numerous data is often relatively time-consuming or expensive. Moreover, the problems of industrial interest tend to be more and more complex and depending on a high number of parameters. High dimensional problems intrinsically involve the need of large data amount through the curse of dimensionality. That is why, new approaches based on smart sampling techniques are investigated to minimize the number of samples to be given to train the model, such as Active Learning methods. Here, we propose a technique based on a combination of Fisher information matrix and of Sparse Proper Generalized Decomposition that enables the definition of a new Active Learning informativeness criterion in high dimensions. We provide examples proving the performances of this technique on a theoretical 5D polynomial function and on an industrial crash simulation application. The results prove that the proposed strategy over-perform the usual ones.

Use of fingerprints found at a crime scene is a common practice for identifying suspects in criminal investigations. Over the past two decades, attempts have been made to obtain additional information from fingerprints, beyond locating suspects as part of an investigation. This includes gender, age and nationality. Researchers demonstrated 75%-90% accuracy in gender classification based on fingerprint images. Nonetheless, despite promising results, these studies have several significant shortcomings with respect to their practical feasibility. First, they ignore the low quality and quantity of fingerprints collected from the crime scene since typically the scene of a crime has only one fingerprint collected, and the fingerprint might be partially or of poor quality as well. Second, as most results are based on a single database, public or private, it is difficult to generalize the most suitable method. Third, studies miss the untapped potential of Data-Centric AI (DCAI) approaches for improving results. The aim of this study was to compare, for the first time, the gender classification from a fingerprint using several datasets and with varying fingerprint image quality. The results from four databases are compared, three public and one internal private database. In addition, we utilize the latest Data-Centric AI (DCAI) approaches for improving classification results. The results demonstrate that a conservative Convolutional Neural Networks (CNN) such as VGG is sufficient for this task. Classification accuracy ranges from 80% to 95% depending on the quality of the fingerprint, with DCAI approaches adding 1%-4% improvement. For partially or low-quality fingerprint images, the periphery of a fingerprint is the most significant area for determining gender. The source code is also provided here for practical application.

It is time-consuming for acquiring complete data by fully phase encoding in two orthogonal directions along with one frequency encoding direction. Undersampling in the 3D k-space is promising in accelerating such 3D MRI process. Though 3D undersampling can be conducted according to predefined probability density, the density based method is not optimal. Because of the large amount of 3D data and computational cost, it is challenging to perform data-driven and learning-based 3D undersampling and subsequent 3D reconstruction. To tackle this challenge, this paper proposes a deep neural network called EEUR-Net, realized by optimizing specific undersampling patterns for the fully-sampled 3D k-space data. Innovatively, our undersampling algorithm employs an end-to-end deep learning approach to optimize phase encoding patterns and uses a 3D U-Net for image reconstruction of undersampled data. Through end-to-end training, we obtain an optimized 3D undersampling pattern, which significantly enhances the quality of the reconstructed image under the same acceleration factor. A series of experiments on a knee MRI dataset demonstrated that, in comparison to standard random uniform, radial, Poisson and equispaced Cartesian undersampling schemes, our end-to-end learned undersampling pattern considerably improves the reconstruction quality of undersampled MRI images.

In Southeast Asia, the incidence of Leukemia, a malignant blood cancer originating from hema-topoietic progenitor cells, is on the rise, marked by a concerning 54% mortality rate. This study focuses on enhancing early-stage prediction to improve patient recovery prospects significantly. Leveraging Machine Learning and Data Science, we employ protein sequential data from frequently mutated genes such as BCL2, HSP90, PARP, and RB to predict Chronic Myeloid Leukemia (CML). Our approach relies on robust feature extraction techniques, namely Di-peptide Composition (DPC), Amino Acid Composition (AAC), and Pseudo amino acid composition (Pse-AAC), with prior attention to addressing outliers and validating feature selection through the Pearson Corre-lation Coefficient. Data augmentation ensures a well-rounded dataset for analysis. Employing a range of Machine Learning models, including Support Vector Machine (SVM), XGBoost, Random Forest (RF), K Nearest Neighbor (KNN), Decision Tree (DT), and Logistic Regression (LR), we achieve accuracy rates spanning from 66% to 94%. These classifiers undergo comprehensive as-sessment using performance metrics such as accuracy, sensitivity, specificity, F1-score, and the confusion matrix. Our proposed solution, encompassing a user-friendly web application dashboard, presents an invaluable tool for early CML diagnosis with profound implications for practitioners, offering a deploy-able asset within healthcare institutions and hospitals.

The realm of natural language processing has always been fascinated by the intricacies of sentence semantics. The emergence of context-aware word representations, especially from pre-trained models like ELMO and BERT, has revolutionized several semantic tasks including but not limited to, question answering, text categorization, and sentiment analysis. Despite these advancements, the integration of supplementary knowledge, particularly syntactic, to augment semantic comprehension in models remains an area ripe for exploration. This paper introduces the Syntactic Enhancement Network (SEN), a pioneering method for synergizing syntactic elements with established pre-trained language models. Our approach encompasses a dual-phase evaluation: initially, we delve into the syntactic augmentation of both RNN-based and Transformer-based language models; subsequently, we test our SEN's proficiency in two specific domains: the task of sentence completion and the extraction of biological relationships. The results are striking, with SEN achieving a stellar 91.2% accuracy in sentence completion—surpassing baseline models by a substantial 37.8%. In the context of biological relation extraction, SEN demonstrates competitive prowess with a 75.1% F1 score.

Our daily activities hinge on the flexibility of our fingers, and a fractured finger can significantly disrupt these routines. The finger bones enable us to bend and fold the fingers and thumb to pick up or grasp objects and do all of our daily activities. A broken finger can cause adverse effects on our daily life activities. It is important to treat broken finger as soon as possible. Swift and precise treatment begins with capturing multiple X-rays, followed by the critical step of fracture detection in these images. Relying on the naked eye for this task carries the risk of overlooking small fractures. To address this issue an automated diagnoses of fractured fingers from images is required for which the current research employs advanced deep learning models—ResNet34, ResNet50, ResNet101, ResNet152, VGG-16, and VGG-19—to classify finger images from the Musculoskeletal Radiographs (MURA) dataset as either fractured or non-fractured. The results emphasize the consistently strong performance of ResNet models, attaining an impressive accuracy of 81.9%. This surpasses VGG models by 3.4% and establishes ResNet as the new benchmark for state-of-the-art accuracy.

When implementing a control system for a ground-based unmanned vehicle (UV) used in agri-culture, an important factor is the low cost of the equipment used, so the use of cameras without lidars for navigation, avoiding obstacles, and ensuring movement safety seems promising. En-suring the safety of movement of UV consists of developing methods and means that make it possible to bypass static and dynamic obstacles and make an emergency stop if it is impossible to bypass. The article describes methods and software that provide detection of objects surrounding the UV through the use of the YOLOv8 neural network, space free for movement through the use of se-mantic segmentation with the MobileNetV3 network architecture, tools for combining several data sources to build a local perception map, as well as linear and angular speed of the UV based on calculating the optimal direction of movement for avoiding obstacles. Testing of the developed methods and software was realized in the Webots simulation environment. Software is developed with ROS2.

Chronic Kidney Disease is a medical condition that causes the decrease in the kidney function and can eventually derive in a total cessation of the work of the organ. It currently affects >10% of the global population, and the number is expected to grow within the next few years, since the correlated diabetes condition is escalating too. Diagnosing the disease on its early stages is crucial to improve life quality, and to increase the chances of survival. The traditional diagnostic methods, which include a biopsy, are invasive, expensive and dangerous. With the use of machine learning algorithms like neural networks, random forest and genetic algorithms, this paper seeks to discover an algorithm with high accuracy, whose selected attributes will be the most optimal combination aiming to reduce the cost, level of invasiveness, and easiness to obtain.

Chinese calligraphy, revered globally for its therapeutic and mindfulness benefits, encompasses styles such as Regular (Kai Shu), Running (Xing Shu), Official (Li Shu), and Cursive (Cao Shu) scripts. Beginners often start with Regular script, advancing to more intricate styles like Cursive. Each style, marked by unique historical calligraphers' contributions, requires learners to discern distinct nuances. The integration of AI in calligraphy analysis, collection, recognition, and classification are pivotal. This study introduces an innovative Convolutional Neural Network (CNN) architecture, pioneering the application of CNN in the classification of Chinese calligraphy. Focusing on the four principal calligraphers' styles from the Tang dynasty (690-907 A.D), this research spotlights the era when the traditional regular script font (Kai Shu) was refined. A comprehensive dataset of 8282 samples from these calligraphers, representing the zenith of regular style, was compiled for CNN training and testing. The model distinguishes personal styles for classification, showing superior performance over existing networks. Achieving 89.5-96.2% accuracy in calligraphy classification, our approach underscores the significance of CNN in both font and artistic style categorization. This research paves the way for advanced studies in Chinese calligraphy and its cultural implications.

In recent years, there has been significant growth in the ubiquity and popularity of three dimensional(3D) point clouds, with an increasing focus on the classification of 3D point clouds. To extract richer features from point clouds, many researchers have turned their attention to various point set regions and channels within irregular point clouds. However, this approach has limited capability in attending to crucial regions of interest in 3D point clouds and may overlook valuable information from neighboring features during feature aggregation. Therefore, this paper proposes a novel 3D point cloud classification method based on global attention and adaptive graph convolution. The method consists of two main branches: the first branch com-putes attention masks for each point, while the second branch employs adaptive graph convolu-tion to extract global features from the point set. It dynamically learns features based on point interactions, generating adaptive kernels to effectively and precisely capture diverse relationships among points from different semantic parts. Experimental results demonstrate that the proposed model achieves 93.8% in overall accuracy and 90.8% in average accuracy on the ModeNet40 dataset.

This paper presents a model for archival and retrieval of the videos of natural flowers. To design an efficient video retrieval system the stages namely, keyframe selection, feature extraction, feature dimensionality reduction and indexing are essential for fast browsing and accessing of videos. Three different keyframe selection approaches are proposed using clustering algorithms after segmenting flower regions from its background. Deep Convolutional Neural Network is used as a feature extractor. After keyframe selection, a video is represented with a set of keyframes. To reduce the feature dimension of a video, two feature selection methods are utilized. For an efficient archival and fast retrieval of flower videos an indexing method called KD-tree is recommended. For a given query video, similar videos are retrieved both in relative and absolute search modalities. An extensive experimentation conducted on a relatively large flower video dataset. The data set consists of 7788 videos of 30 different species of flowers. The videos are captured with three different devices in different resolutions. The comparative study reveals proposed keyframe selection approaches gives better results. It has also been observed that the videos retrieved in absolute approach with features selected from Binormal separation metric and indexing gives good results.

Skin cancer is one of the widespread diseases that typically develop on the skin due to continuous exposure to sunlight. Although cancer can appear on any part of the human body, skin cancer reports account for over half of all cancer occurrences worldwide. There are substantial obstacles to the precise diagnosis and classification of skin lesions because of the morphological variety and indistinguishable characteristics across skin malignancies. Recently, Deep Learning models have been used in the field of image-based lesion diagnosis, and it has demonstrated diagnostic efficiency on par with that of dermatologists. To increase classification efficiency and accuracy for skin lesions, a cutting-edge multi-layer deep Convolutional Neural Network (CNN) termed SkinLesNet has been built in this study. The ResNetV50 and VGG16 models have been carefully compared to review the performance of the proposed model. The dataset used in this study, PAD-UFES-20, contains 1314 samples in total and includes three common forms of skin lesions. The proposed approach, SkinLesNet, significantly outperforms the well-known compared models in the given conditions

The development and use of highly automated vehicles (HAV) in agriculture is a highly important task because of the lack of machine drivers and unsatisfactory level of their qualifications, reducing the impact of harmful factors on employees of agricultural enterprises. HAVs allow to avoid these problems: increase productivity and the efficiency of seasonal use of equipment, and reduce the impact of harmful factors on people. Modern HAVs in agriculture are based on the following methods: parallel driving according to geo coordinates without automatic turning and without obstacles tracking, and on technical vision without the use of a geographic information system (for example Cognitive AgroPilot) and a digital field map, which makes it impossible to generate the optimal path around the field.

This paper presents the Advanced Syntactic-Graph Convolutional Model (ASGCM), a pioneering approach in Aspect-Based Sentiment Analysis (ABSA) that integrates syntactic dependency features within a graph convolution framework. ASGCM stands out for its novel use of dependency edge encoding and tag-based graph convolutions, providing a fine-grained analysis of sentiments associated with specific aspects in text. This model meticulously captures the intricacies of syntactic structures, thereby offering enhanced precision in sentiment analysis. Notably, ASGCM incorporates a dual-layer graph convolution system: one layer processes syntactic dependencies (edges), while the other interprets semantic roles (tags), ensuring a comprehensive understanding of both structural and contextual elements in text. We rigorously tested ASGCM on multiple datasets, including both English and Chinese languages, and our findings reveal a significant improvement in sentiment classification accuracy compared to existing models. The versatility of ASGCM makes it a robust tool for diverse linguistic environments, setting a new standard for ABSA methodologies.

This paper presents an approach to enhancing the clarity and intelligibility of speech in digital communications compromised by various background noises. Utilizing deep learning techniques, specifically a Variational Autoencoder (VAE) with 2D convolutional filters, we aim to suppress background noise in audio signals. Our method focuses on four simulated environmental noise scenarios: storms, wind, traffic, and aircraft. Training dataset has been obtained from public sources (TED-LIUM 3 dataset, which includes audio recordings from the popular TED-TALK series) combining with these background noises. The audio signals were transformed into 2D power spectrograms, upon which our VAE model was trained to filter out the noise and reconstruct clean audio. Our results demonstrate that the model outperforms existing state-of-the-art solutions in noise sup-pression. Although differences in noise types were observed, it was challenging to definitively conclude which background noise most adversely affects speech quality. Results have been assessed with objective methods (mathematical metrics) and subjective (listening to a set of audios by humans). Notably, wind noise showed the smallest deviation between the noisy and cleaned audio, perceived subjectively as the most improved scenario. Future work involves refining the phase calculation of the cleaned audio and creating a more balanced dataset to minimize differences in audio quality across scenarios. Additionally, practical ap-plications of the model in real-time streaming audio are envisaged. This research contributes significantly to the field of audio signal processing by offering a deep learning solution tailored to various noise conditions, enhancing digital communication quality.

The groundbreaking Advanced Forest-Based Tagging Framework (AFTF) represents a paradigm shift in the domain of information extraction from medical texts. AFTF introduces a novel dual-binary tree structure that redefines how entity-relation triples are extracted. This innovative approach directly tackles the limitations of traditional linear and graph-based methods, effectively addressing challenges related to overlapping triples and computational efficiency. The AFTF model stands as a beacon of excellence, surpassing established baselines by significant margins in comprehensive evaluations on two pivotal medical datasets. Notably, AFTF achieves remarkable improvements in F1 scores, showcasing its prowess in accurate information extraction from complex medical narratives. Beyond its exceptional performance in the medical domain, AFTF exhibits remarkable versatility. This adaptability is vividly demonstrated through its robust performance across three diverse public datasets, further affirming its position as a versatile and reliable solution for information extraction tasks. This paper provides a comprehensive exposition of the AFTF architecture, shedding light on its innovative design principles and its efficient handling of intricate medical texts. AFTF represents a groundbreaking step forward in the realm of information extraction, promising enhanced accuracy, efficiency, and adaptability for a wide range of applications.

In this paper, we present a groundbreaking methodology in the realm of aspect-level sentiment analysis, which capitalizes on the advanced capabilities of graph-based neural networks. Our approach, distinguished as the Aspect Correlation Graph Network (ACGN), represents a significant departure from conventional models. These traditional models often analyze aspects in isolation, failing to capture the intricate web of sentiment relationships that may exist within a single sentence. ACGN, however, is designed to address this gap by employing a sophisticated bidirectional attention mechanism, integrated with positional encoding. This unique combination not only enhances the model's ability to focus on relevant parts of the sentence but also aids in constructing detailed, aspect-focused representations. These representations are particularly crucial for understanding the nuanced interplay of sentiments associated with different aspects. Central to our model's architecture is the incorporation of a graph convolutional network. This network serves as a pivotal component in mapping and analyzing the complex network of sentiment correlations that can exist among various aspects within sentences. Through this integration, ACGN is able to unearth and interpret the subtle and often overlooked sentiment dynamics that traditional models might miss. Our comprehensive evaluations of the Aspect Correlation Graph Network, conducted using the SemEval 2014 datasets, have yielded promising results. These findings demonstrate a clear and significant advancement over the capabilities of existing models. Particularly, the results underscore the critical importance and utility of recognizing and understanding the connections between sentiments of different aspects in text analysis. This insight opens new avenues in the field of sentiment analysis, suggesting a broader application potential of ACGN in various contexts where understanding nuanced sentiment relationships is key. Overall, our study not only introduces a novel approach in aspect-level sentiment analysis but also sets a new standard for future research in this area. By highlighting the integral role of inter-aspect sentiment connections, ACGN paves the way for more sophisticated and accurate sentiment analysis tools, capable of handling the complexities of natural language with greater finesse and precision.

Equipping autonomous agents for dynamic interaction and navigation is a significant challenge in intelligent transportation systems. This study aims to address this by implementing a brain-inspired model for decision-making in autonomous vehicles. We employ active inference, a Bayesian approach that models decision-making processes similar to the human brain, focusing on the agent's preferences and the principle of free energy. This approach is combined with imitation learning to enhance the vehicle's ability to adapt to new observations and make human-like decisions. The research involved developing a multi-modal self-awareness architecture for autonomous driving systems and testing this model in driving scenarios, including abnormal observations. The results demonstrated the model's effectiveness in enabling the vehicle to make safe decisions, particularly in unobserved or dynamic environments. The study concludes that the integration of active inference with imitation learning significantly improves the performance of autonomous vehicles, offering a promising direction for future developments in intelligent transportation systems.

A key component for the proper functioning, availability, and reliability of power grids is the power trans- former. Although these are very reliable assets, the early detection of incipient degradation mechanisms are very important to prevent failures that may shorten their lifetime. In this work a review and comparative analysis, of classical Machine Learning algorithms (such as single and ensemble classification algorithms) and two automatic machine learning classifiers, are presented for the fault diagnosis of power transformers. The goal is to determine whether fully automated ML approaches are worse or better than traditional ML frameworks that require a human in the loop (such as a data scientist) to identify transformer faults from oil-dissolved gases data. The methodology uses a DB compiled from published and proprietary transformer fault samples. Faults data is obtained from the literature, the Duval pentagon method, and user-expert knowledge. The parameters from, either single or ensemble classifiers, were optimized through standard machine learning procedures. The results showed that the best performing algorithm is a robust automatic machine learning classifiers model followed by classical algorithms such as neural networks and stacking ensembles. These results highlight the ability of a robust automatic machine learning model for, handling imbalanced power transformers faults datasets with high accuracy, using the minimum tuning effort by the electrical experts. By providing the most probable transformer fault condition will reduce the time required to find and solve the fault.