The speech entailed in human voice comprises essentially para-linguistic information used in many voice-recognition applications. Gender voice-recognition is considered one of the pivotal parts to be detected from a given voice, a task that involves certain complications. In order to distinguish gender from a voice signal, a set of techniques have been employed to determine relevant features to be utilized for building a model from a training set. This model is useful for determining the gender (i.e, male or female) from a voice signal. The contributions are involved in two folds: (i) providing analysis information about well-known voice signal features using a prominent dataset, (ii) studying various machine learning models of different theoretical families to classify the voice gender, and (iii) using three prominent feature selection algorithms to find promisingly optimal features for improving classification models. Experimental results show the importance of sub-features over others, which are vital for enhancing the efficiency of classification models performance. Experimentation reveals that the best recall value is equal to 99.97%; 99.7% of two models of Deep Learning (DL) and Support Vector Machine (SVM) and with feature selection the best recall value is 100% for SVM techniques.

Speech/music classification that facilitates optimized signal processing from classification results has been extensively adapted as an essential part of various electronics applications, such as multi-rate audio codecs, automatic speech recognition, and multimedia document indexing. In this paper, a new technique to improve the robustness of speech/music classifier for 3GPP enhanced voice service (EVS) using long short-term memory (LSTM) is proposed. For effective speech/music classification, feature vectors implemented with the LSTM are chosen from the features of the EVS. Experiments show that LSTM-based speech/music classification produces better results than conventional EVS under a variety of conditions and types of speech/music data.

This paper presents a novel approach to sentiment analysis specifically customized for predicting stock market movements, bypassing the need for external dictionaries which are often unavailable for many languages. Our methodology directly analyzes textual data, with a particular focus on context-specific sentiment words within neural network models. This specificity ensures our sentiment analysis is both relevant and accurate in identifying trends in the stock market. We employ sophisticated mathematical modeling techniques to enhance both the precision and interpretability of our models. Through meticulous data handling and advanced machine learning methods, we leverage large datasets from Twitter and financial markets to examine the impact of social media sentiment on financial trends. We achieved an accuracy exceeding 75%, highlighting the effectiveness of our modeling approach, which we further refined into a convolutional neural network model. This achievement contributes valuable insights into sentiment analysis within the financial domain, thereby improving the overall clarity of forecasting in this field.

This study focuses on understanding how global auto parts companies manage innovation by analyzing keywords in CEO messages from their reports. The study involved three key phases: (1) extraction of innovation-related keywords emphasized in CEO messages, (2) comparison of innovation-related keyword usage between high-performing (upper group) and low-performing (lower group) companies, and (3) analysis of the characteristics of innovation management via internal meaning and the structural interpretation of keywords. A comparison between the upper and lower groups revealed distinct differences in the emphasized keywords. The mapping of open innovation success factors, based on keywords extracted according to eigenvector centrality, identified Provision of Resources and Governance as central to innovation management in the upper group. These findings serve as a benchmark for establishing the direction, vision, and strategy of innovation management in the lower group.

This text mining study delves into the multifaceted contributions of the energy sector to the Sus-tainable Development Goals (SDGs). By analyzing a wide range of academic literature, we un-cover key themes, trends, and challenges shaping the intersection of energy and sustainability. The findings reveal that the energy sector plays a pivotal role in achieving SDGs such as affordable and clean energy (SDG 7) and climate action (SDG 13). Critical issues encompass governance, policy frameworks, and technological innovations. This research underscores the need for inter-disciplinary collaboration and holistic approaches in addressing complex energy-related sus-tainability challenges. The insights derived here provide guidance to policymakers, researchers, and stakeholders seeking to harness the energy sector's potential for a more sustainable and eq-uitable future.

In the realm of the five-category classification endeavor, there has been limited exploration into applied techniques for classifying Arabic text. These methods have primarily leaned on singletask learning, incorporating manually crafted features that lack robust sentence representations. Recently, the Transformer paradigm has emerged as a highly promising alternative. However, when these models are trained using single-task learning, they often face challenges in achieving outstanding performance and generating robust latent feature representations, especially when dealing with small datasets. This issue is particularly pronounced in the context of the Arabic dialect, which has a scarcity of available resources. Given these constraints, this study introduces an innovative approach to dissecting sentiment in Arabic text. This approach combines Inductive Transfer (INT) with the Transformer paradigm to augment the adaptability of the model and refine the representation of sentences. By employing self-attention SE-A and feed-forward sub-layers as a shared Transformer encoder for both the five-category and three-category Arabic text classification tasks, this proposed model adeptly discerns sentiment in Arabic dialect sentences. The empirical findings underscore the commendable performance of the proposed model, as demonstrated in assessments of the Hotel Arabic-Reviews Dataset, the Book Reviews Arabic Dataset, and the LARB dataset.

Subway construction is often in a complex natural and human-machine operating environment, and that complicated setting leads to subway construction more prone to safety accidents, which can cause substantial casualties and monetary losses. Thus, it is necessary to investigate the safety risks of subway construction. The existing literature on the identification and assessment of subway construction safety risks(SCSR) is susceptible to the influence of subjective factors. Moreover, although existing studies have explored the interrelationships between different risks, these studies usually analyze the interrelationships of single risks, lack the study of risk chain transfer relationships, and fail to find out the key path of risk transfer. Therefore, this paper innovatively combines text mining, association rules and complex networks to deep mine subway construction safety incident reports and explore risk transfer process. Firstly, it uses text mining technology to identify subway construction safety risk; Then, association rules are introduced to explore the causal relationships among safety risk; Finally, the key safety risk and important transfer paths of subway construction safety accidents (SCSA) are obtained based on the complex network model. Research results show that (a) improper safety management, unimplemented safety subject responsibilities, violation of operation rules, non-perfect safety responsibilities system and insufficient safety education and training are the key safety risk in SCSA; (b) two shorter key risk transfer paths in the subway construction safety network can be obtained: insufficient safety education and training→lower safety awareness→violation of operation rules→safety accidents; insufficient safety checks or hidden trouble investigations→violation of operation rules→safety accidents; (c) in the process of risk transfer, the risk can be controlled by controlling the key nodes or cutting off the transfer path. The results of the study provide new ideas and methods for SCSR identification and influence element mining, which help safety managers propose accurate subway construction safety risk control measures.

The article aims to explore social aspects in the area of energy research by using bibliometrics methods and tools. The study is based on the data of Energy Research & Social Science journal publications exported from ScienceDirect database during 2019-2023. Methodologically, the work included clustering of the bibliometric records under consideration. This allowed to identify the main topics of publications and those in which the research of social problems prevails. This stage of the work was carried out using carrot2, an engine for clustering search results. Next, the analysis of publication topics related to social issues research was carried out using VOSviewer — a software tool for visualizing scientific landscapes. The topics of the four clusters obtained at this stage were described by key terms with the maximum total link strength. For the principal ones, a definition was made on the basis of the texts of publications that most fully disclose the meaning of these terms. The 1537 publications in the journal during the period under review deal with many aspects of energy transition research, renewable energy, energy policy, energy justice and energy systems, but only 78 of them are dominated by social issues. The paper reveals the main social issues in the field of energy research: sociology of energy, socio-technical system, community energy, energy democracy, sociology of the future, energy sovereignty, socio-technical imaginary and shows the insufficient attention to this topic, especially to the investment of such research.

Requirements are predominantly written in Natural Language (NL), which makes them accessible to stakeholders with varying degrees of experience, as compared to a model-based language which requires special training. However, despite its benefits, NL can introduce ambiguities and inconsistencies in requirements, which may eventually result in system quality degradation and system failure altogether. The system complexity that characterizes current systems warrants an integrated and comprehensive approach to system design and development. This need has brought about a paradigm shift towards Model-Based Systems Engineering (MBSE) approaches to system design and a departure from traditional document-centric methods. While, MBSE shows great promise, the ambiguities and inconsistencies present in NL requirements hinder their conversion to models directly. The field of Natural Language Processing (NLP) has demonstrated great potential in facilitating the conversion of NL requirements into a semi-machine-readable format that enables their standardization and use in a model-based environment. A first step towards standardizing requirements consists of classifying them according to the ``type'' (design, functional, performance, etc.) they represent. To that end, a language model capable of classifying requirements needs to be fine-tuned on labeled aerospace requirements. This paper presents the development of an annotated aerospace requirements corpus and its use to fine-tune BERT$_\text{BASE-UNCASED}$ to obtain aeroBERT-Classifier: a new language model for classifying aerospace requirements into design, functional, or performance requirements. A comparison between aeroBERT-Classifier and bart-large-mnli (zero-shot text classification) showcased the superior performance of aeroBERT-Classifier on classifying aerospace requirements despite being fine-tuned on a small labeled dataset.

Introduction: Around a million people are reported to die by suicide every year, and due to the stigma associated with the nature of the death, this figure is usually assumed to be an underestimate. Suicide may be prevented if prompt intervention is taken to mitigate risk. Machine learning and artificial intelligence-based modelling, such as natural language processing (NLP) and other text analytics approaches, has the potential to become a major technique for the detection, diagnosis, and treatment of people who are suffering from mental health issues. The primary aims of this research are to determine whether NLP techniques have been utilised in the field of suicide prevention, and if so, were they effective? What were their limitations? Methods: PubMed, EMBASE, MEDLINE, PsycInfo, and Global Health databases were searched for studies that reported use of NLP for suicide ideation or self-harm. Thematic analysis was used to synthesise and analyse the included studies. Findings were reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement, and the Mixed Methods Appraisal Tool (MMAT) was used in assessing paper quality. Result: The preliminary search of five databases generated 387 results. Removal of duplicates resulted in 158 potentially suitable studies. Twenty papers were finally included in this review. Discussion: Studies show that combining structured and unstructured data in NLP data modelling yielded more accurate results than utilizing either alone. Also, to reduce suicides, people with mental problems must be continuously and passively monitored. Further, NLP and other machine learning/artificial intelligence technologies can be used to address health inequities and electronic health records provide valuable data for creating suicide risk tools. Finally, Online, social media, and smartphone applications can be leverage in detecting people with suicide ideation. Conclusion: The use of artificial intelligence and machine learning opens new avenues for considerably guiding risk prediction and advancing suicide prevention frameworks. The review's analysis of the included research revealed that the use of NLP may result in low-cost and effective alternatives to existing resource-intensive methods of suicide prevention. To summarise, there is substantial evidence that NLP is useful in identifying people who have suicide ideation.

Nowadays, an analysis of virtual media to predict society’s reaction to any events or processes is a task of great relevance. Especially it concerns meaningful information on healthcare problems. Internet sources contain a large amount of pharmacologically meaningful information useful for pharmacovigilance purposes and repurposing drug use. An analysis of such a scale of information demands developing the methods that require the creation of a corpus with labeled relations among entities. Before, there have been no such Russian language datasets. This paper considers the first Russian language dataset where labeled entity pairs are divided into multiple contexts within a single text (by used drugs, by different users, by the cases of use, etc.), and a method based on the XLM-RoBERTa language model, previously trained on medical texts to evaluate the state-of-the-art accuracy for the task of indication of the four types of relationships among entities: ADR–Drugname, Drugname–Diseasename, Drugname–SourceInfoDrug, Diseasename–Indication. As shown based on the presented dataset from the Russian Drug Review Corpus, the developed method achieves the F1-score of 81.2% (obtained using cross-validation and averaged for the four types of relationships), which is 7.8% higher than the basic classifiers.

Study question: What are the current trends of research in Human Assisted Reproduction around the world? Summary answer: USA is the leading country, followed by the UK, China, France and Italy. The largest research area is “laboratory techniques”, although other areas such as “public health”, “quality, ethics and law” and “female factor” are gaining ground worldwide. What is known already: Scientific research, especially in health and medical sciences, aims at addressing specific needs that society (and, especially, patients) perceives as pressing. One of the main challenges for policymakers and research funders alike is therefore to align research priorities to societal needs. We can thus think of research agendas in terms of a demand side (societal needs) and a supply side (research outputs). Research output in Human Assisted Reproduction has expanded in the past years, as indicated by the increasing number of scientific publications in indexed journals in this area. Nevertheless, no map of research related to assisted reproduction has been produced so far, hindering the identification of potential areas of improvement and need. Study design, size, duration: 26,000+ scientific publications (articles, letters, and reviews) on Human Assisted Reproduction produced worldwide between 2005 and 2016 were analyzed. These publications were indexed in PubMed or obtained from reference list of indexed publications included in the analysis.Participants/materials, setting, methods: The corpus of publications was obtained by combining the MeSH terms: “Reproductive techniques”, “Reproductive medicine”, “Reproductive health”, “Fertility”, “Infertility”, and “Germ cells”. Then it was analyzed by means of text mining algorithms (Topic Modeling (TM) based on Latent Dirichlet Allocation (LDA)), in order to obtain the main topics of interest. Finally, these categories were analyzed across world regions and time. Main results and the role of chance: We identified 44 main topics, which were further grouped in 11 macro categories, form larger to smaller: “laboratory techniques”, “male factor”, “quality, ethics and law”, “female factor”, “public health and infectious diseases”, “basic research and genetics”, “pregnancy complications and risks”, “general infertility and ART”, “psychosocial aspects”, “cancer”, and “research methodology”. The USA was the leading country in number of publications, followed by the UK, China, France and Italy. Interestingly, research contents in high income countries is fairly homogeneous across macro-categories, and it is dominated by “laboratory techniques” in Western and Southern Europe, and by “quality, ethics and law” in North America, Australia and New Zealand. In middle income countries we observe that research is mainly performed on “male factor”, and noticeably less on “female factor”. Finally, research on “public health and infectious diseases” predominates in low-income countries. Regarding temporal evolution of research, “laboratory techniques” is the most abundant topic on a yearly basis, and relatively constant over time. However, since production in most of the other categories is increasing, the relative contribution of this research category is actually decreasing. Publication is especially increasing in “public health and infectious diseases” (in all world regions, but especially in low income countries), “quality, ethics and law” (high income countries), and “female factor” (middle income countries). Limitations, reasons for caution: Three main factors might limit the robustness of our work: the textual corpus analyzed is based on abstract and titles, the reproducibility of the stochastic algorithms applied, which may produce slightly differing results at each run, and the interpretation of the topics obtained. Wider implications of the findings: This study should prove beneficial in the design of research strategies and policies that foster the alignment between supply (assisted reproduction research) and demand (society). Study funding/competing interest(s): PTQ-14-06718 of the Spanish MINECO Torres Quevedo programme (FAM).

Text fields in electronic medical records (EMR) contain information on important factors that influence health outcomes, however, they are underutilized in clinical decision making due to their unstructured nature. We analyzed 6,497 inpatient surgical cases with 719,308 free text notes from Le Bonheur Children’s Hospital EMR. We used a text mining approach on preoperative notes to obtain the text-based risk score algorithm as predictive of death within 30 days of surgery. We studied the additional performance obtained by including text-based risk score as a predictor of death along with other structured data based clinical risk factors. The C-statistic of a logistic regression model with 5-fold cross-validation significantly improved from 0.76 to 0.92 when text-based risk scores were included in addition to structured data. We conclude that preoperative free text notes in EMR include significant information that can predict adverse surgery outcomes.

At present, medical institutes generally use EMR to record patient's condition, including diagnostic information, procedures performed and treatment results. EMR has been recognized as a valuable resource for large scale analysis. However, EMR has the characteristics of diversity, incompleteness, redundancy and privacy, which make it difficult to carry out data mining and analysis directly. Therefore, it is necessary to preprocess the source data in order to improve data quality and improve the data mining results. Different types of data require different processing technologies. Most structured data commonly needs classic preprocessing technologies, including data cleansing, data integration, data transformation and data reduction. For semi-structured or unstructured data, such as medical text, containing more health information, it requires more complex and challenging processing methods. The task of information extraction for medical texts mainly includes NER (Named Entity Recognition) and RE (Relation Extraction). In this paper, we introduce the process of EMR processing, including data collection, data preprocessing, data mining, evaluation and knowledge application, analyze the current status of the key technologies, such as data preprocessing and data mining, and provide an overview of the application domains and prospects of EMR mining technologies. Finally, we summarize the existing problems in the research of EMR mining, and review the development trends.

The research examines the relationship between self-rated health situation and personal percep-tion of heat waves, and how social linkage of communities would be a moderator variable in residents’ perception of heat waves in Taiwan. This study uses the questionnaire conducted by Sinica “Responsive Capacity under Heat Wave: The Perspectives of the Locals”(2019), using OLS method for estimating the unknown parameters in multiple regression model. The author finds that the correlation of self-rated health situation and perception toward heat is significantly posi-tive. Also, social linkage in communities affects strongly as a moderator variable: While the sat-isfaction to with their community could reduce negative reaction to heat, contacts with neighbors could increase possibility people feel uncomfortable in high-temperature situation. This study ex-hibits the effects of social environment on community, and expects further related researches or practices to strengthen capability to resist heat wavesƒ for Taiwanese residents.

Amebiasis is a disease caused by the unicellular parasite Entamoeba histolytica. In most cases, the infection is asymptomatic but when symptomatic, the infection can cause dysentery and invasive extraintestinal complications. In the gut, E.histolytica feeds on bacteria. Increasing evidences support the role of the gut microbiota in the development of the disease. In this review we will discuss the consequences of E.histolytica infection on the gut microbiota. We will also discuss new evidences about the role of the gut microbiota in regulating the resistance of the parasite to oxi-dative stress and its virulence.

Objective: To explore three novel fasting blood glucose (FBG)-based novel indicators, including FBG-to-albumin ratio (FAR), FBG-to-lymphocytes ratio (FLR), and FBG-to-hemoglobin ratio (FHR), in predicting prognosis and detecting metastasis for patients with pancreatic neuroendocrine neoplasms after resection. Materials and methods: A total of 178 pNEN patients underwent surgical resection were included in this study. Receiver operating characteristic (ROC) curves were used to evaluate the diagnosis values of FAR, FLR, and FHR, and the cut-off values were obtained for further analyses. Univariate and multivariate analyses were conducted to determine the independent predictors. The Kaplan-Meier method was used to evaluate the progression-free survival (PFS) and overall survival (OS) of the pNEN patients. Results: The optimal cut-off values of FAR, FLR, and FHR were 0.17, 2.85, and 0.028, respectively. As for PFS, the area under the curve (AUC) was 0.693 for FAR, 0.690 for FLR, and 0.661 for FHR, respectively. The AUC was 0.770, 0.692, and 0.715 accordingly for OS. The groups with lower FAR, FLR, and FHR were significantly associated with prolonged PFS and OS (p < 0.05). In patients with metastasis, the lower FAR group was correlated with significantly longer PFS and OS (p = 0.022 and 0.002, respectively). The FLR was an independent predictor of PFS in pNEN patients, and the FAR was a predictor of OS. FAR was an independent indicator of PFS in patients with metastasis. Conclusion: Preoperative FAR, FLR, and FHR are effective in predicting the prognosis of pNEN patients and detecting the synchronous metastases.

Objective: The current study aimed to delineate the relationship of furin with inflammation while cervical intraepithelial neoplasia progresses to cancer. Study Design:This cross-sectional study included 81 women who required colposcopic examinations. The study groups were formed based on pathological results: Group I included women with cervical intraepithelial neoplasia (CIN) I (n=30); Group II included women with CIN II-III (n=28); and Group III included women with cervical cancer (CC) (n=23). Furin, ki-67, and p16 levels were evaluated based on immunostaining intensity. The inflammatory indices were calculated parallel with the literature from routine blood samples retrieved within one week before the procedure. Results:Furin expression gradually increased from CIN I to CIN II-III and from CIN II-III to CC, respectively (p < .001, p = .005). The NLR, MLR, PLR, and SII were significantly higher in the CC group (p < .001). ROC curve analysis unveiled that NLR, MLR, PLR and SII predicted the presence of CC with a cutoff value of 2.39 for NLR (sensitivity: 91.3%, specificity: 63.8%, AUROC: 0.79, p < .001); a cutoff value of 0.27 for MLR (sensitivity: 78.3%, specificity: 72.4%, AUROC: 0.77, p = .009); a cutoff value of 123 for PLR (sensitivity: 100%, specificity: 41.4%, AUROC: 0.70, p = .04); and a cutoff value of 747 for SII (sensitivity: 69.6%, specificity: 90.7%, AUROC: 0.71, p = .014). Conclusion:Furin expression increased gradually in parallel to the severity of cervical intraepithelial neoplasia. The inflammatory indices were higher in the presence of CC and denoted good discrimination ability for predicting cervical cancer.

Purpose: School-age children with autism spectrum conditions (ASC) often experience difficulties in speech-in-noise (SiN) perception, leading to increased listening effort that impacts their well-being and academic performance. This study aimed to investigate the SiN processing challenges faced by Mandarin-speaking children with ASC and its impact on their listening effort. Methods: Participants completed sentence recognition tests in both quiet and noisy conditions, with a steady-state noise masker presented at 0 dB signal-to-noise ratio in the noisy condition. We compared recognition accuracy and task-evoked pupil responses from 23 Mandarin-speaking children with ASC to 19 age-matched neurotypical (NT) counterparts to gauge their behavioral performance and listening effort during these auditory tasks. Results: The ASC group demonstrated notably decreased accuracy in noise compared to their NT peers, suggesting poorer SiN perception. Pupillometric data further revealed significantly larger peak dilations in the ASC group than in the NT group under comparable conditions. Importantly, the ASC group's peak dilation in quiet mirrored the NT group's in noise. However, the ASC group exhibited shorter peak latencies and reduced mean dilations than the NT group in similar conditions. Such patterns indicate the ASC group might initially experience a heightened cognitive load but utilize fewer cognitive resources as the task continued, indicating an atypical allocation of cognitive resources and a potential tendency towards relatively superficial and automated auditory processing. Conclusion: Our findings highlight the unique SiN processing challenges children with ASC face, underscoring the importance of a nuanced, individual-centric approach for interventions and support.

Electric stimulation via a Cochlear Implant (CI) enables people with severe to profound sensorineural hearing loss to regain speech understanding and music appreciation and thus allowing them to actively engage in social life. Three main manufacturers (Cochlear, MED-EL and Advanced Bionics “AB”) have been offering CI systems, thus challenging CI recipients and Otolaryngologists with a difficult decision, as currently no comprehensive overview or meta-analyses on performance outcome following CI implantation is available. The main goal of this scoping review is to provide evidence that data and standardized speech and music performance tests are available for performing such comparisons. To this end, a literature search was conducted to find studies that address speech and music outcomes in CI recipients. From a total of 1592 papers, 188 paper abstracts were analyzed and 147 articles were found suitable for examination of full text. From which, 42 studies were included for synthesis. A total of 16 studies used the consonant-nucleus-consonant (CNC) word recognition test in quiet at 60db SPL. We found that aside from technical comparisons, only very few publications compare speech outcomes across manufacturers of CI systems. Evidence suggests though, that these data are available in large CI centers in Germany and US. Future studies should therefore leverage large data cohorts to perform such comparisons that could provide critical evaluation criteria and assist both CI recipients and Otolaryngologists to make informed performance-based decisions.

Radiological imaging is an essential component of a swallowing assessment. Artificial intelligence (AI), and especially deep learning (DL) models, have enhanced the efficiency and efficacy through which imaging is interpreted, and subsequently have important implications for swallow diagnostics and intervention planning. However, the application of AI for the interpretation of videofluoroscopic swallowing studies (VFSS) is still emerging. This review showcases recent literature in the use of AI to interpret VFSS and highlights clinical implications for speech pathologists (SPs). With a surge in AI research, there have been advances made in dysphagia assessment. Several studies have demonstrated successful implementation of DL algorithms to analyze VFSS. Notably, convolutional neural networks (CNNs) have been used to detect pertinent aspects of the swallowing process with high levels of precision. DL algorithms have the potential to streamline VFSS interpretation, improve efficiency and accuracy, and enable precise interpretation of instrumental dysphagia evaluation, which is especially advantageous when access to skilled clinicians is not ubiquitous. By enhancing precision, speed, and depth of VFSS interpretation, SPs can obtain a more comprehensive understanding of swallow physiology and deliver targeted and timely intervention that is tailored towards the individual. This has practical applications for both clinical practice and dysphagia research. As this research area grows and AI technologies progress, the application of DL in the field of VFSS interpretation is clinically beneficial and has the potential to transform dysphagia assessment and management. With broader validation and inter-disciplinary collaborations, AI-augmented VFSS interpretation is likely to transform swallow evaluation and ultimately improve outcomes for individuals with dysphagia.

Background: In the present study, we assessed the feasibility and success outcomes of voice prosthesis (VP) changes when performed by speech-language pathologist (SLP). Methods: Patients treated with total laryngectomy (TL) from January 2020 to December 2020 were prospectively recruited from our medical center. Patients benefited from tracheoesophageal puncture. The VP changes were performed by the senior SLP and the following data were collected for each VP change: date of placement; change or removal; VP type and size; reason for change or removal; and use of a washer for periprosthetic leakage. A patient-reported outcome questionnaire including 6 items was proposed to patients at each VP change (Appendix 1). Items were assessed with a 10-point Likert-scale. Results: Fifty-two VP changes were performed by the senior SLP during the study period. The mean duration of the SLP consultation, including patient history, examination and VP change procedure was 20 min (range: 15-30). The median prosthesis lifetime was 88 days. The main reasons for VP changes were transprosthetic (N=34; 79%) and periprosthetic (N=7; 21%) leakages, respectively. SLP successfully performed all VP changes. He did not change one VP but used a periprosthetic silastic to stop the periprosthetic leakages. In two cases, SLP needed the surgeon examination to discuss about the following indication: implant mucosa inclusion and autologous fat injection. The patient satisfaction was high according to the speed and the quality of care by the SLP. Conclusion: The delegation of VP change from the otolaryngologist-head and neck sur-geon to the speech-language pathologist (SLP) may be done without significant complications. The delegation of VP change procedure to SLP may be interesting in some rural regions with otolaryngologist shortage.

Observing customers is one of the methods to uncover their needs. By closely observing how customers use products, we can indirectly experience their interactions and gain a deep understanding of their feelings and preferences. Through this process, companies can design new products that have the potential to succeed in the market. However, traditional methods of customer observation are time-consuming and labor-intensive. In this study, we propose a method that leverages the analysis of online customer reviews as a substitute for direct customer observations. By correlating Customer Journey Map (CJM) with online reviews, this research establishes a verb-centric analysis that produces a CJM based on online review data. Various Text analysis techniques were utilized in this process. Through empirical data analysis, we validated that indirect observation based on customer reviews can be performed more efficiently than traditional methods. Additionally, we observed that the customer behavior VOC (Voice of Customer) identified during the CJM mapping process offers unique insights that are distinct from traditional product feature-centric review analyses. To verify the usefulness of these behavior VOC, we evaluated the quantitative analysis results of the same reviews through product development experts. Experts recognized these customer behavior VOC as useful information for developing new products.

Sentiment analysis plays a crucial role in understanding customer feedback, guiding product development, and informing business decisions. This paper introduces term sentiment entropy (TSE), a novel weighting method that leverages the distribution of sentiment labels associated with words to enhance text classification accuracy. TSE complements traditional TF-IDF techniques by capturing the nuances of sentiment variation across different contexts. Experiments across diverse public datasets demonstrate TSE's potential to improve sentiment analysis performance, especially in capturing subtle sentiment shifts and adapting to specific domains. While computational cost and label quality pose challenges, TSE offers a promising avenue for refining sentiment analysis and opening new research frontiers.

For highly time-constrained very short-term investors, reading and extracting valuable information from financial news poses significant challenges. The wide range of topics covered in these news articles further compounds the difficulties for investors. The diverse content adds complexity and uncertainty to the text, making it arduous for very short-term investors to swiftly and accurately extract valuable insights. Variations in authors, media sources, and cultural backgrounds also introduce additional complexities. Hence, performing a bull-bear semantic analysis of financial news using text mining technologies can alleviate the volume, time, and energy pressures on very short-term investors while enhancing the efficiency and accuracy of their investment decisions. This study proposes labeling bull-bear words from a futures corpus detection method that extracts valuable information from financial news, allowing investors to understand market trends quickly. Generative AI models are trained to provide real-time bull-bear advice, aiding investors in adapting to market changes and devising effective trading strategies. Experimental results show the effectiveness of various models, with Random Forest and SVMs achieving an impressive 80% accuracy rate. MLP and Deep learning models also perform well. By leveraging these models, the study reduces the time spent reading financial articles, enabling faster decision-making and increasing the likelihood of investment success. Future research can explore the application of this method in other domains and enhance model design for improved predictive capabilities and practicality.

With the increasing use of cloud-based services for training and deploying machine learning models, data privacy has become a major concern. This is particularly important for natural language processing (NLP) models, which often process sensitive information such as personal communications and confidential documents. In this study, we propose a method for training NLP models on encrypted text data to mitigate data privacy concerns while maintaining similar performance to models trained on non-encrypted data. We demonstrate our method using two different architectures, namely Doc2Vec+XGBoost and Doc2Vec+LSTM, and evaluate the models on the 20 Newsgroups dataset. Our results indicate that both encrypted and non-encrypted models achieve comparable performance, suggesting that our encryption method is effective in preserving data privacy without sacrificing model accuracy. In order to replicate our experiments, we have provided a Colab notebook at the following address: https://t.ly/lR-TP

In this paper, we introduce a model for the automatic generation of literary sentences in French. It is based on algorithms that we have previously used to generate sentences in Spanish and Portuguese, and on a new corpus consisting of literary texts in French that we have constructed, called [FR]. Our automatic text generation algorithm combines language models, shallow parsing and deep learning, artificial neural networks. We have also proposed and implemented a manual evaluation protocol to assess the quality of the artificial sentences generated by our algorithm, by testing if they fulfill four simple criteria. We have obtained encouraging results from the evaluators for most of the desired features of our artificially generated sentences.

The ability to identify and rank risk is essential for efficient and effective supervision of financial service firms, such as banks and insurers. Risk ranking ensures limited resources are allocated where they are most needed. Today, automatic risk identification within insurance supervision primarily relies on quantitative metrics based on numerical data (e.g. returns). The purpose of this work is to assess whether Natural Language Processing (NLP) and cognitive networks can achieve similar automated risk ranking and identification by analysing textual data, i.e. NIDT=829 investor transcripts from Bloomberg. To this aim, this work explores and tunes 3 NLP techniques: (1) keyword extraction enhanced by cognitive network analysis; (2) valence/sentiment analysis; and (3) topic modelling. Results highlight that keyword analysis, enriched by term frequency-inverse document frequency scores and semantic framing through cognitive networks, could detect events of relevance for the insurance system like cyber-attacks or the COVID-19 pandemic. Cognitive networks were found to highlight events that related to specific financial transitions: The semantic frame of "climate" grew in size by +538% between 2018 and 2020 and outlined an increased awareness that agents and insurers expressed towards climate change. A lexicon-based sentiment analysis achieved a Pearson’s correlation of ρ=0.16 (p&lt;0.001,N=829) between sentiment levels and daily share prices. Although relatively weak, this finding indicates that insurance jargon is insightful to support risk supervision. Topic modelling is considered less amenable to support supervision, because of a lack of results’ stability and an intrinsic difficulty to interpret risk patterns. We discuss how these automatic methods could complement existing supervisory tools in automated risk ranking.

In the current business environment, firms are eager to adopt new technology as they witness and perceive more and more successful business applications of new technologies, e.g., Big Data, Artificial Intelligence (A.I.), Cloud Computing, etc. As one of the disruptive technologies, Blockchain technology (BCT) is now drawing public attention owing to the cryptocurrency phenomenon (e.g., Bitcoin), for which Blockchain serves as the backbone technology. Given certain innovative features of BCT, especially its transparency, traceability, security, efficiency, confidentiality, and immutability, BCT holds out the promise of impacting supply chain operational and financial efficiencies. Recently, the burgeoning of Metaverse and Non-Fungible Token (NFT) has boosted the BCT as state-of-the-art technology to another notch. Motivated by the proliferating adoption of BCT, we conduct a holistic literature review with a focus on the status of research on BCT in the context of Supply Chain Management (SCM). In particular, this Blockchain-centered research reviews the research up to date on the Blockchain application in SCM. It provides holistic review in terms of (1) the functionality of BCT and its salient features; (2) the prevailing and potential applications of BCT; (3) and the business benefits and impact of BCT in SCM. Finally, we substantiate and highlight a variety of research directions.

Thanks to rapid development of artificial intelligence technology in recent years, the current artificial intelligence technology is contributing to many part of society. Education, environment, medical care, military, tourism, economy, politics, etc. are having a very large impact on society as a whole. For example, in the field of education, there is an artificial intelligence tutoring system that automatically assigns tutors based on student's level. In the field of economics, there are quantitative investment methods that automatically analyze large amounts of data to find investment laws to create investment models or predict changes in financial markets. As such, artificial intelligence technology is being used in various fields. So, it is very important to know exactly what factors have an important influence on each field of artificial intelligence technology and how the relationship between each field is connected. Therefore, it is necessary to analyze artificial intelligence technology in each field. In this paper, we analyze patent documents related to artificial intelligence technology. We propose a method for keyword analysis within factors using artificial intelligence patent data sets for artificial intelligence technology analysis. This is a model that relies on feature engineering based on deep learning model named KeyBERT, and using vector space model. A case study of collecting and analyzing artificial intelligence patent data was conducted to show how the proposed model can be applied to real-world problems.

In this paper, we propose a novel approach of video skimming by exploiting the fusion of video temporal information and keyword information representation extracted from multi-model video information including audio, text and visual indices. In addition, we introduce the brand-safe filtering and sentiment analysis in order to only reserve the user-friendly content in the video skim. In the experiment by using the videos from YouTube-8M dataset, we have proved that the semantic conservation in the video skim from the proposed approach highly outperforms the approaches by only partial information of the video in conserving the semantic content of the video.

(1) Background: Little research has been done on professionals' perceptions of institutions and governments during epidemics. We aim to create a profile of physicians who feel they can raise public health issues with relevant institutions during a pandemic. (2) Methods: A total of 1285 Romanian physicians completed an online survey as part of a larger study. We used binary logistic regression to profile physicians who felt able to raise public health issues with relevant institutions; (3) Results: Five predictors could differentiate between respondents who tended to agree with the trust statement and those who tended to disagree: feeling safe at work during the pandemic, considering the financial incentive worth the risk, receiving training on the use of protective equipment, having the same values as colleagues, and enjoying work as much as before the pandemic; (4) Conclusions: Physicians who trust the system to raise public health issues with the appropriate institutions feel they share the same values as their colleagues, say they were trained in the use of protective equipment used during the pandemic, felt they were safe at work during the pandemic, enjoyed their work as much as before the pandemic, and felt the financial bonus justified the risk.

The properties of the annual, semiannual and triennial oscillations (AO, SAO and TO) in the middle atmosphere have been investigated using the TIMED/SABER temperature data. The Lomb-Scargle and wavelet spectra were used to determine the dominant oscillations in the background temperature field. The AO is prominent at the mid-latitudes. The AO amplitudes present an asymmetry between the two Hemispheres, being larger in the mesosphere than in the stratosphere. The SAO dominates the tropical regions, with three amplitude maxima at altitudes of 45, 75, and 85 km. The SAOs in the upper mesosphere (75 km) are out of phase with those in the mesopause (85 km) in the tropical regions, which can generate an enhancement of 11 K at each equinox, contributing to the lower mesospheric inversion layer. The TO is significant in the tropical region, with amplitude being maximum at 35, 45 and 85 km. Result shows that there may be potential interaction by the TO with SAO at 85km at the equator. The relation between ENSO and TO has also been discussed. The ENSO signal may modulate the amplitude of the TO, mainly in the lower stratosphere. The real origin of the TO may lie in the wave-mean-flow interaction.

Economic development and population growth lead to increased production of chicken manure (CM), which is a problematic organic waste for its amount, environmental threats, and moisture content. There are different ways of CM, namely anaerobic digestion, composting, combustion, and direct land spreading. Hydrothermal carbonization (HTC) is another emerging way, however. In this study, the HTC of CM was performed to produce energy-rich material called hydrochar (HC). The effects of HTC temperature (180, 240, 300 C) and process time (30, 90, 180 min) were summarized. Proximate and ultimate analysis, as well as heating values (HHV, LHV), have been performed both on raw CM and derived HC. Additionally, the process performance has been examined. The obtained results show that HTC is a feasible method for CM disposal and valorization. Although process time did not influence considerably fuel properties of CM, higher temperature led to significantly higher HHV, reaching 23,880.6734.56 Jg-1 at 300 C and 180 min with an improvement of 8,329 Jg-1 compared with raw CM (15,551.67 Jg-1). The process conducted at 240 C in 30 min has been specified as the most favorable, due to the highest energy gain of HC and relatively low energy consumption.

The increase in the emission of greenhouse gases (GHG) is one of the most important world problems. Decreasing of GHG emission is a big challenge in the future. Transportation sector uses a significant part of petroleum production in the world and it leads to an increase in the emission of GHG. The result of this issue is that population of the world befoul environment of transportation system automatically. Electric Vehicles (EV) have a potentiality to solve a big part of the GHG emission and energy efficiency issues such as stability and reality of energy. The energy actors and their research team determine some targets for 2050; hence, they hope to decrease the world temperature to 6 °C in the best and 2 °C in the normal condition. Fulfillment of these scenarios needs suitable grid infrastructure but in most of the countries, the grid does not have a suitable background to apply those scenarios. In this paper, some problems about energy scenarios, energy storage systems, grid infrastructure and communication systems in supply and demand side of the grid and its solutions have been investigated.

Vehicle to vehicle (V2V) and Vehicle to infrastructure (V2I) or briefly V2X communications are the one of hot topics in automotive industry. Therefore, this situation is providing many advantages of connected vehicles and infrastructures which bring to human life. For instance, vehicles and road infrastructures which shares information with each other, provides a neat flow regulation, more ordered traffic flow and therefore jammed traffic dependent accident’s percentage will be decreased. On the other hand, security is the most important issue for these systems because the operation of V2X networks is completely dependent on uninterrupted and accurate information sharing. In the light of these information, in this paper we review security issues and current solution architectures. We also propose some open problems in this lively field.

Sustaining learners’ motivation to keep learning has been concerned for over 60 years in education and SLA. Most research focusing on Japanese university students has contributed lots of theoretical and practical developments, but research focusing on motivations of non-traditional adult learners and those under 18-year-old to learn Chinese is still little even though previous research findings have proved age is one of the key variables influencing learning besides nationalities and Chinese language levels. In order to find an effective treatment for sustainable Chinese language learning in Japan, this study utilized a survey method to analyze the types of Japanese learners’ motivation to learn Chinese via analysis of a moment structures (AMOS), and to compare the differences of the motivation via one-way ANOVA. This study revealed that Japanese learners’ motivation to learn Chinese consists of eight common types, which are “instrumental motivation”, “personal orientation”, “identified regulation”, “Chinese cultural productions”, “integrating into Chinese community”, “external regulation”, “social responsibility”, and “Chinese for academic purposes”. Also, there are similarities and differences existing in the eight motivation types among the Japanese children, adolescents, (non-) traditional adult learners of Chinese as a foreign language (CFL). The findings contribute the understanding of motivation types and differences among the four aged Japanese learners to teachers of teaching Chinese to speakers of other languages (TCSOL), and to future research further exploring how differently aged Japanese learners can have and use a higher motivation as a main drive to learn Chinese language as a lifelong business.

A community-level nutrition intervention was implemented among tribal children (3 to 6 years) in Telangana, India. The one-year intervention involved six nutrient-rich formulations of millet-pulse-groundnut based products, suited to local taste preferences. Anthropometric measurements of height, weight and Mid-Upper-Arm -Circumference (MUAC) along with haemoglobin (Hb) levels were monitored at baseline and endline. The treatment group showed considerable gain in height (3.2 cm), weight (1.68 kg) and MUAC (0.33 cm), over the control group. Paired t-test indicated significant differences (p<.01) between pre- and post-intervention anthropometric measurements. Positive shifts were observed in wasting (WHZ; -1.2 ±1.3 to -0.9± 1), stunting (HAZ; -1.8±1.6 to -0.3±1.3) and underweight (WAZ; -1.9±1.2 to -0.7±1) among treatment group. Hb levels in the treatment group also improved significantly from 9.70 ± 0.14 g/dl (moderately-anaemic) to 11.08 ± 0.13 g/dl (non-anaemic). Post-intervention Focus Group Discussions (FGDs) involving mothers and teachers confirmed the positive impact. Thus, nutritional intervention formulated using climate resilient millets, pulses and groundnut, promotes dietary diversity and improves the nutrition & health status of children.

Talker variability has been reported to facilitate generalization and retention of speech learning, but is also shown to place demands on cognitive resources. Our recent study provided evidence that phonetically-irrelevant acoustic variability in single-talker (ST) speech is sufficient to induce equivalent amounts of learning to the use of multiple-talker (MT) training. This study is a follow-up contrasting MT versus ST training with varying degrees of temporal exaggeration to examine how cognitive measures of individual learners may influence the role of input variability in immediate learning and long-term retention. Native Chinese-speaking adults were trained on the English /i/-/ɪ/ contrast. We assessed the trainees’ working memory and selective attention before training. Trained participants showed retention of more native-like cue weighting in both perception and production regardless of talker variability condition. The ST training group showed long-term benefit in word identification, whereas the MT training group did not retain the improvement. The results demonstrate the role of phonetically-irrelevant variability in robust speech learning and modulatory functions of nonlinguistic working memory and selective attention, highlighting the necessity to consider the interaction between input characteristics, task difficulty, and individual differences in cognitive abilities in assessing learning outcomes.

Artificial Intelligence is increasingly being used by social media platforms to tackle online hate speech. The sheer quantity of content, the speed at which is it developed and the enhanced pressure companies are facing by States to remove hate speech quickly from their platforms have led to a tricky situation. This commentary argues that automated mechanisms, which may have biased datasets and be unable to pick up on the nuances of language, should not be left unattended with hate speech as this can lead to issues of violating freedom of expression and the right to non-discrimination.

The real challenge in Human Robot Interaction (HRI) is to build machines capable of perceiving human emotions so that robots can interact with humans in a proper manner. It is well known from the literature that emotion varies accordingly to many factors. Among these, gender represents one of the most influencing one, and so an appropriate gender-dependent emotion recognition system is recommended. In this paper, a two-level hierarchical Speech Emotion Recognition (SER) system is proposed: the first level is represented by the Gender Recognition (GR) module for the speaker’s gender identification; the second is a gender-specific SER block. Specifically for this work, the attention was focused on the optimisation of the first level of the proposed architecture. The system was designed to be installed on social robots for hospitalised and living at home elderly patients monitoring. Hence, the importance of reducing the software computational effort of the architecture also minimizing the hardware bulkiness, in order for the system to be suitable for social robots. The algorithm was executed on the Raspberry Pi hardware. For the training, the Italian emotional database EMOVO was used. Results show a GR accuracy value of 97.8%, comparable with the ones found in literature.

Many aging adults experience some form of hearing problems that may arise from auditory peripheral damage. However, it has been increasingly acknowledged that hearing loss is not only a dysfunction of the auditory periphery but results from changes within the entire auditory system, from periphery to cortex. Damage to the auditory periphery is associated with an increase in neural activity at various stages throughout the auditory pathway. Here, we review neurophysiological evidence of hyperactivity, auditory perceptual difficulties that may result from hyperactivity, and outline open conceptual and methodological questions related to the study of hyperactivity. We suggest that hyperactivity alters all aspects of hearing – including spectral, temporal, spatial hearing – and, in turn, impairs speech comprehension when background sound is present. By focusing on the perceptual consequences of hyperactivity and the potential challenges of investigating hyperactivity in humans, we hope to bring animal and human electrophysiologists closer together to better understand hearing problems in older adulthood.

We propose a novel transfer learning method for speech emotion recognition allowing us to obtain promising results when only few training data is available. With as low as 125 examples per emotion class, we were able to reach a higher accuracy than a strong baseline trained on 8 times more data. Our method leverages knowledge contained in pre-trained speech representations extracted from models trained on a more general self-supervised task which doesn’t require human annotations, such as the wav2vec model. We provide detailed insights on the benefits of our approach by varying the training data size, which can help labeling teams to work more efficiently. We compare performance with other popular methods on the IEMOCAP dataset, a well-benchmarked dataset among the Speech Emotion Recognition (SER) research community. Furthermore, we demonstrate that results can be greatly improved by combining acoustic and linguistic knowledge from transfer learning. We align acoustic pre-trained representations with semantic representations from the BERT model through an attention-based recurrent neural network. Performance improves significantly when combining both modalities and scales with the amount of data. When trained on the full IEMOCAP dataset, we reach a new state-of-the-art of 73.9% unweighted accuracy (UA).

This paper propose an original approach of achieving a Cymatics based visual perception of isolated speech commands. The idea is to smartly combine the effective speech processing and analysis methods with the phenomena of Cymatics. In this context, an effective approach for automatic isolated speech based message recognition is proposed. The incoming speech segment is enhanced by applying the appropriate pre-emphasis filtering, noise thresholding and zero alignment operations. The Mel-Frequency Cepstral coefficients (MFCCs), Delta coefficients and Delta-Delta coefficients are extracted from the enhanced speech segment. Later on, the Dynamic Time Warping (DTW) technique is employed to compare these extracted features with the reference templates. The comparison outcomes are used to make the classification decision. The classification decision is transformed into a methodical excitation. Finally, this excitation is converted into the systematic visual perceptions via the phenomenon of Cymatics. The system functionality is tested with an experimental setup and results are presented. The approach is novel and can be employed in various applications like visual art, encryption, education, archeology, architecture, integration of impaired people, etc.

This paper proposes a novel approach, based on the adaptive rate processing and analysis, for the isolated speech recognition. The idea is to smartly combine the event-driven signal acquisition and windowing along with adaptive rate processing, analysis and classification for realizing an effective isolated speech recognition. The incoming speech signal is digitized with an event-driven A/D converter (EDADC). The output of EDADC is windowed with an activity selection process. These windows are later on resampled uniformly with an adaptive rate interpolator. The resampled windows are de-noised with an adaptive rate filter and their spectrum are computed with an adaptive resolution short time Fourier transform (ARSTFT). Later on, the magnitude, Delta and Delta-Delta spectral coefficients are extracted. The Dynamic Time Warping (DTW) technique is employed to compare these extracted features with the reference templates. The comparison outcomes are used to make the classification decision. The system functionality is tested for a case study and results are presented. An 8.2 times reduction in acquired number of samples is achieved by the devised approach as compared to the classical one. It aptitudes a significant computational gain and power consumption reduction of the proposed system over the counter classical ones. An average subject dependent isolated speech recognition accuracy of 96.8% is achieved. It shows that the proposed approach is a potential candidate for the automatic speech recognition applications like rehabilitation centers, smart call centers, smart homes, etc.

This study aimed to identify changes in visitor behavior and visitor interest in healing forests before and after the COVID-19 pandemic. The study used text mining analysis techniques to identify changes in visitation behavior over time, divided into three periods: pre-COVID-19 (January 1 to December 31, 2019), during the COVID-19 pandemic (November 1, 2020 to October 31, 2022), and post-COVID-19 (November 1, 2022 to October 31, 2023 ). After the COVID-19 outbreak, healing forest use behavior did not regress to pre-COVID-19 levels. Activity-based keywords such as "hiking," "trekking," and "walking" stood out as the main drivers of this change in behavior. Therefore, related authorities must examine the scalability of the functions, services, and programs of healing forests from a general healing space to a space for leisure and tourism. These findings will contribute to the development of future marketing strategies and programs for healing forests.

Binarization of degraded documents is an important preprocessing task for various document analysis such as OCR and historical document analysis. Existing studies have applied various convolutional neural network (CNN) models and generative models for document binarization, but they do not show generalized performance for noise that the model has not seen and it suffers from extracting elaborate text strokes. In this paper, to overcome these challenges, we utilize latent diffusion model (LDM), which is known for high-quality image generation model, for the first time in document binarization. By utilizing the iterative diffusion-denoising process in latent space, it shows high-quality cleaned binarized image generation and high generalized performance through using both data distribution and time step while training. Additionally, we apply gated U-Net to the backbone network to preserve text strokes using trainable gating value. Gated convolution can extract elaborate text stroke by allowing the model to focus on text region by combining gating value and feature. Furthermore, we maximize the effectiveness of the proposed model by training it with a combination of LDM loss and pixel-level loss, which is suitable for the model structure. Experiments on H-DIBCO and DIBCO benchmark datasets show that the proposed model outperforms existing methods.

This paper considers the reasonability of using GSDMM as a method for clustering short texts - titles and abstracts of publications 2021-2023 of MDPI journals on sustainable energy topics. The paper proposes an approach to identifying relevant research topics based on the use of the Python script Yake!, the Krovetz streamer, the GSDMM algorithm, and short text annotations. It is emphasized that researchers prefer to rely on specific publications rather than keywords when searching for information on a topic of interest. The bibliometric records of sustainable energy publications from 2021 to 2023 in Sustainability (Energy Sustainability section — 1,926 search results) and Energies (Sustainable Energy section — 994 search results) were used as data for analysis. The GSDMM algorithm was used to cluster the texts, and Yake! to extract the keywords for the GSDMM algorithm's vocabulary. This article summarizes topics found in 9 clusters, including general sustainable energy issues, biomass recycling, region-specific issues, building energy use, economic growth and investment, numerical flow modeling, generation cost optimization, heat to energy conversion, weather-related risks, and false data attacks.

Industrial nameplates serve as a means of conveying critical information and parameters. In this work, we propose a novel approach for rectifying industrial nameplate pictures utilizing a probabilistic Hough transform. Our method effectively corrects for distortions and clipping, and features a collection of challenging nameplate pictures for analysis. To determine the corners of the nameplate, we employ a progressive probability Hough transform, which not only enhances detection accuracy but also possesses the ability to handle complex industrial scenarios. The results of our approach are clear and readable nameplate text, as demonstrated through experiments that show improved accuracy in model identification compared to other methods.

This paper investigates if Corporate Social Responsibility (CSR) reports published by a selected group of Nordic companies are aligned with the Global Reporting Initiative (GRI) standards. To achieve this goal, several natural language processing, and text mining techniques were implemented and tested. We extracted strings, corpus, and hybrid semantic similarities from the reports and evaluated the models through the intrinsic assessment methodology. A quantitative ranking score based on index matching was developed to complement the semantic valuation. The final results show that Latent Semantic Analysis (LSA) and Global Vectors for Word Representation (GloVE) are the best methods for our study. Our findings will open the door to the automatic evaluation of sustainability reports which could have a strong impact on the environment.

For any molecule, network, or process of interest, to keep up with new publications on these, is becoming increasingly difficult. For many cellular processes, molecules and their interactions that need to be considered can be very large. Automated mining of publications can support large scale molecular interaction maps and database curation. Text mining and Natural Language Processing (NLP)-based techniques are finding their applications in mining the biological literature, handling problems such as Named Entity Recognition (NER) and Relationship Extraction (RE). Both rule-based and machine learning (ML)-based NLP approaches have been popular in this context, with multiple research and review articles examining the scope of such models in Biological Literature Mining (BLM). In this review article, we explore self-attention based models, a special type of neural network (NN)-based architectures that have recently revitalized the field of NLP, applied to biological texts. We cover self-attention models operating either at a sentence level or an abstract level, in the context of molecular interaction extraction, published from 2019 onwards. We conduct a comparative study of the models in terms of their architecture. Moreover, we also discuss some limitations in the field of BLM that identifies opportunities for the extraction of molecular interactions from biological text.

Ethnopharmacology experts face several challenges when identifying and retrieving documents and resources related to their scientific focus. The volume of sources that need to be monitored, the variety of formats utilized, the different quality of language use across sources, present some of what we call “big data” challenges in the analysis of this data. This study aims to understand if and how experts can be supported effectively through intelligent tools in the task of ethnopharmacological literature research. To this end, we utilize a real case study of ethnopharmacology research, aimed at the Southern Balkans and Coastal zone of Asia Minor. Thus, we propose a methodology for more efficient research in ethnopharmacology. Our work follows an “Expert-Apprentice” paradigm in an automatic URL extraction process, through crawling, where the apprentice is a Machine Learning (ML) algorithm, utilizing a combination of Active Learning (AL) and Reinforcement Learning (RL), and the Expert is the human researcher. ML-powered research improved 3.1 times the effectiveness and 5.14 times the efficiency of the domain expert, fetching a total number of 420 relevant ethnopharmacological documents in only 7 hours versus an estimated 36-hour human-expert effort. Therefore, utilizing Artificial Intelligence (AI) tools to support the researcher can boost the efficiency and effectiveness of the identification and retrieval of appropriate documents.

This article presents a dataset of tweets in the Urdu language. There are 1,140,824 tweets in the dataset, collected from Twitter for September and October 2020. This large-scale corpus of tweets is generated by performing pre-processing which includes removing columns containing user information, retweet’s count, followers information, duplicate tweets, removing unnecessary punctuation, links, symbols, and spaces, and finally extracting emojis if present in the tweet text. In the final dataset each tweet record contains columns for tweet id, text, and emoji extracted from the text with a sentiment score. Emojis are extracted to validate Machine Learning models used for the multilingual sentiment and behavior analysis. These are extracted using a Python script that searches for an emoji from the list of 751 most frequently used emojis. If an emoji is present in the text, a column with the emoji description and sentiment score is added.

The fact is quite transparent that almost everybody around the world is using android apps. Half of the population of this planet is associated with messaging, social media, gaming, and browsers. This online marketplace provides free and paid access to users. On the Google Play store, users are encouraged to download countless of applications belonging to predefined categories. In this research paper, we have scrapped thousands of users reviews and app ratings. We have scrapped 148 apps’ reviews from 14 categories. We have collected 506259 reviews from Google play store and subsequently checked the semantics of reviews about some applications form users to determine whether reviews are positive, negative, or neutral. We have evaluated the results by using different machine learning algorithms like Naïve Bayes, Random Forest, and Logistic Regression algorithm. we have calculated Term Frequency (TF) and Inverse Document Frequency (IDF) with different parameters like accuracy, precision, recall, and F1 and compared the statistical result of these algorithms. We have visualized these statistical results in the form of a bar chart. In this paper, the analysis of each algorithm is performed one by one, and the results have been compared. Eventually, We've discovered that Logistic Regression is the best algorithm for a review-analysis of all Google play store. We have proved that Logistic Regression gets the speed of precision, accuracy, recall, and F1 in both after preprocessing and data collection of this dataset.

Recent state-of-the-art scene text recognition methods have primarily focused on horizontal text in images. However, in several Asian countries, including China, large amounts of text in signs, books, and TV commercials are vertically directed. Because the horizontal and vertical texts exhibit different characteristics, developing an algorithm that can simultaneously recognize both types of text in real environments is necessary. To address this problem, we adopted the direction encoding mask (DEM) and selective attention network (SAN) methods based on supervised learning. DEM contains directional information to compensate in cases that lack text direction; therefore, our network is trained using this information to handle the vertical text. The SAN method is designed to work individually for both types of text. To train the network to recognize both types of text and to evaluate the effectiveness of the designed model, we prepared a new synthetic vertical text dataset and collected an actual vertical text dataset (VTD142) from the Web. Using these datasets, we proved that our proposed model can accurately recognize both vertical and horizontal text and can achieve state-of-the-art results in experiments using benchmark datasets, including the street view test (SVT), IIIT-5k, and ICDAR. Although our model is relatively simple as compared to its predecessors, it maintains the accuracy and is trained in an end-to-end manner.

Access to medicines is one of the essential problems in Public Health of low- and middle-income countries (LMICs). The World Health Organization (WHO) defines access to medicines as the possibility of "having continuously accessible and affordable medicines in public or private health facilities that are within a kilometer of the place of residence." Access to medicines, as defined by the WHO, is not fully guaranteed in many LMICs and even in many regions of high-resource countries. The WHO identifies several factors as determinants of limitations in the access to medicines: rational selection, affordable prices, sustainable financing, and reliable health services. The action on these factors makes it possible to improve universal access to medicines with consequent improvement in Public Health. Adequate access to antibiotics and vaccination will avoid a large part of the deaths caused by infectious diseases in the LMICs. However, the emergence of resistance and the difficulties in vaccination campaigns due to socio-political or cultural problems make it challenging to fight many easily treatable infectious diseases. The use and abuse of antibiotics are inevitably associated with the appearance of resistances that make them ineffective. Thus, whereas limited access to antibiotics raises mortality rates from infectious diseases, generalized open access to them ends up eliminating their clinical value. Moreover, the contraction of research in this field for many years has reduced the success in discovering new drugs. Additionally, local market regulations, inadequate selection, inaccessible prices, especially for those of second and third-generation, inefficient health systems, and difficulties of administration and control of prescription compliance, especially in the case of combined therapies, are additional obstacles to universal access to antibiotics. In order to simultaneously improve access to antibiotics and keep resistances under control, it is necessary to develop training and education activities at different social levels (from patients to various Health Care Providers) to complement the national or supranational strategic plans.

Neuropsychiatric manifestations of COVID-19 include delirium, general malaise, headache, and psychiatric symptoms. These manifestations are related to the neuroinflammatory response to viral antigens and proinflammatory mediators/immune cells. Tetracyclines, such as minocycline, have antibacterial and anti-inflammatory properties. Saiko-keishi-to, a traditional Japanese Kampo medicine, also has anti-inflammatory properties. Given the anti-inflammatory properties of minocycline and Saiko-keishi-to, we describe two cases of COVID-19 with prolonged headaches and general malaise successfully treated with these medications.

This study aims to describe the commitment to national heritage, patriotism, nationalism of elementary school students in the border areas between Indonesia and Timor Leste. This research was conducted in three elementary schools, they are SD Inpres Moto Ain, SD Inpres Class Jau Halimuti, SD Kristen Nanaeklot, and SD Negeri Motabenar from June to August 2018. The subjects of this study were elementary students in the Indonesia-Timor Leste border areas precisely around Motaain City, Belu Regency, East Nusa Tenggara. The research questionnaire was adapted from Karazawa and analyzed with descripriptive statistics. The results of this study showed that the commitment to national heritage, patriotism, and nationalism of elementary students on the Indonesian border areas with Timor Leste is in the very high level.

Aim: To investigate inflammation indices and erythropoietin (EPO) levels for their potential role in distinguishing polycythemia vera (PV) from secondary polycythemia (SP), and to compare different parameter combinations in terms of diagnostic accuracy. Methods: This retrospective cohort was created from patients assessed for polycythemia from January 2020 to December 2023. PV diagnosis was made according to the 2016 WHO criteria (n=145). Those who did not fulfill the criteria were defined as having SP (n=84). Results: NLR, PLR and SII were significantly higher in the PV group (p

The study comprehensively evaluated the prognostic roles of PLR, NLR, MLR, BLR, and ELR in patients with acute exacerbation chronic obstructive pulmonary disease (AECOPD). 619 patients with AECOPD and 300 healthy volunteers were retrospectively included into the study. The clinical characteristics containing laboratory findings of the AECOPD patients and the blood cell counts (CBCs) of the healthy volunteers were collected. Compared with the healthy volunteers, PLR, NLR, and MLR were elevated in COPD patients in stable condition, and were further ele-vated during exacerbation. ELR showed the opposite trend. PLR, NLR, and MLR were all posi-tively correlated with hospital LOS as well as CRP. In contrast, ELR was negatively correlated with hospital LOS as well as CRP. Elevated PLR, NLR, and MLR were all associated with more serious airflow limitation in AECOPD. Elevated PLR, NLR, and MLR were all associated with increased in-hospital mortality while Elevated ELR was associated with decreased in-hospital mortality in AECOPD. A nomogram was construct to predict in-hospital mortality in AECOPD. The nomo-gram had a C-index of 0.850 (95% CI: 0.799 – 0.901) with good predictive value and clinical ap-plicability. In summary, PLR, NLR, MLR, and ELR served as predictors for clinical outcomes in patients with AECOPD.

Nollywood, based on the idea of Bollywood from India, is a series of outstanding movies that originate from Nigeria. Unfortunately, while the movies are in English, they are hard to understand for many native speakers due to the dialect of English that is spoken. In this article, we accomplish two goals: (1) create a phonetic sub-title model that is able to translate Nigerian English speech to American English and (2) use the most advanced toxicity detectors to discover how toxic the speech is. Our aim is to highlight the text in these videos which is often times ignored for lack of dialectal understanding due the fact that many people in Nigeria speak a native language like Hausa at home.

Speech emotion recognition remains a heavy lifting in natural language processing. It has strict requirements to the effectiveness of feature extraction and that of acoustic model. With that in mind, a Heterogeneous Parallel Convolution Bi-LSTM model is proposed to address these challenges. It consists of two heterogeneous branches: the left one contains two dense layers and a Bi-LSTM layer, while the right one contains a dense layer, a convolution layer, and a Bi-LSTM layer. It can exploit the spatiotemporal information more effectively, and achieves 84.65%, 79.67%, and 56.50% unweighted average recall on the benchmark databases EMODB, CASIA, and SAVEE, respectively. Compared with the previous research results, the proposed model achieves better performance stably.

Theories of reading comprehension emphasize decoding and listening comprehension as two essential components. The current study aimed to investigate how Chinese character decoding and context-driven auditory semantic integration contribute to reading comprehension in Chinese middle school students. Seventy-five middle school students were tested. Context-driven auditory semantic integration was assessed with speech-in-noise tests in which the fundamental frequency (F0) contours of spoken sentences were either kept natural or acoustically flattened with the latter requiring a higher degree of contextual information. Statistical modelling with hierarchical regression was conducted to examine the contributions of Chinese character decoding and context-driven auditory semantic integration to reading comprehension. Performance on Chinese character decoding and auditory semantic integration scores with the flattened (but not natural) F0 sentences significantly predicted reading comprehension. Furthermore, the contributions of these two factors to reading comprehension were better fitted with an additive model instead of a multiplicative model. These findings indicate that reading comprehension in middle schoolers is associated with not only character decoding but also the listening ability to make better use of the sentential context for semantic integration in a severely degraded speech-in-noise condition. The results add to our better understanding of the multi-faceted reading comprehension in children. Future research could further address the age-dependent development and maturation of reading skills by examining and controlling other important cognitive variables, and apply neuroimaging techniques such as functional magmatic resonance imaging to reveal the neural substrates for the contribution of auditory semantic integration and the observed additive model to reading comprehension.

Purpose: Dysarthria, a neurological injury of the motor component of the speech circuitry, is of common consequences of traumatic brain injury (TBI). Palilalia is a speech disorder characterized by involuntary repetition of words, phrases, or sentences. Based on the evidence supporting the effectiveness of transcranial direct current stimulation (tDCS) in some speech and language disorders, we hypothesized that using tDCS would enhances the effectiveness of speech therapy in a client with chronic dysarthria following TBI. Method: We applied the constructs of the “Be Clear” protocol, a relatively new approach in speech therapy in dysarthria, together with tDCS on a chronic subject who affected by dysarthria and palilalia after TBI. Since there was no research on the use of tDCS in such cases, regions of interest (ROIs) were identified based on deviant brain electrophysiological patterns in speech tasks and resting state compared with normal expected patterns using the Quantitative Electroencephalography (QEEG) analysis. Results: Measures of perceptual assessments of intelligibility, an important index in the assessment of dysarthria, ​​were superior to the primary protocol results immediately and 4 months after intervention. We did not find any factor other than the use of tDCS to justify this superiority. The percentage of repeated words, an index in palilalia assessment, had a remarkable improvement immediately after intervention but fell somewhat after 4 months. We justified this case with subcortical origins of palilalia. Conclusion: Our present case-based findings suggested that applying tDCS together with speech therapy may improve intelligibility in similar case profiles as compared to traditional speech therapy. To reconfirm the effectiveness of the above approach in cases with dysarthria following TBI, more investigation need to be pursued.

Gesture recognition has always been a technique to decrease the distance between the physical and the digital world. In this work, we introduce an Arduino based vehicle system which no longer require manual controlling of the cars. The proposed work is achieved by utilizing the Arduino microcontroller, accelerometer, RF sender/receiver, and Bluetooth. Two main contributions are presented in this work. Firstly, we show that the car can be controlled with hand-gestures according to the movement and position of the hand. Secondly, the proposed car system is further extended to be controlled by an android based mobile application having different modes (e.g., touch buttons mode, voice recognition mode). In addition, an automatic obstacle detection system is introduced to improve the safety measurements to avoid any hazards. The proposed systems are designed at lab-scale prototype to experimentally validate the efficiency, accuracy, and affordability of the systems. We remark that the proposed systems can be implemented under real conditions at large-scale in the future that will be useful in automobiles and robotics applications.

1) This study describes the good evolution of a 6-year-old girl genetically diagnosed with Rett syndrome (RTT), after having been treated with IGF-1, MT (MT), blackcurrant extracts (BC), and rehabilitation during 6 months. 2) The patient stopped her normal development from the first year of age. The patient showed low weight and height and met the main criteria for typical RTT. Curiously, there was pubic hair (Tanner II), very high plasma testosterone, despite low gonadotropins. No adrenal enzymatic deficits existed, and ultrasound abdominal studies were normal. Treatment consisted in IGF-1 (0.04 mg/kg/day, 5/week, sc) during 3-months and then 15-days resting, MT (50 mg/day, orally, uninterruptedly) and neurorehabilitation. The new blood tests were absolutely normal and the pubic hair disappeared. Then, a new treatment with IGF-1, MT, and BC started for another 3 months. After it, pubic Tanner stage increased to III, without a known cause. 3) The treatment followed led to clear improvements in most of the initial impairments, perhaps because of the effect of IGF-I, the antioxidant effects of MT and BC, and the increase in cyclic-glycine-proline (cGP) after BC administration. 4) A continuous treatment with IGF-1, MT and BC may recover most of the neurologic disabilities that occur in RTT.

YouTube is a very popular social media platform in today’s digital landscape. The primary focus of this study is to explore the underlying sentiment in user comments on food-related videos on YouTube, specifically within two pivotal food product categories: Plant-Based and Hedonic. Our research involved labelling comments using sentiment lexicons such as TextBlob and VADER. Furthermore, the sentiments of the comments were classified using advanced Machine Learning (ML) algorithms, namely Support Vector Machine (SVM), Multinomial Naive Bayes, Random Forest, Logistic Regression, and XGBoost. The evaluation of these models encompassed key macro average metrics, including accuracy, precision, recall, and f1-score. Results from VADER showcased a high accuracy level, with SVM achieving 93% accuracy in the plant-based dataset and 96% in the hedonic dataset. In addition to sentiment analysis, we delved into user interactions within the two datasets, measuring crucial metrics such as views, likes, comments, and engagement rate. The findings illuminate significantly higher levels of views, likes, and comments in the hedonic food dataset, but the plant-based dataset maintains a superior overall engagement rate.

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.

In recent years, the high-frequency abnormal price fluctuations of small-scale agricultural products have been detrimental to market stability. The traditional method based on statistical data of production factors can no longer accurately predict abnormal price fluctuations. Taking scallions as a case, the paper identifies price influencing factors from massive network information, and analyzes the causes and changing characteristics of price fluctuations from four perspectives: market supply, economic environment and market sentiment, and attention. Initially, the study employs an LDA topic model to extract factors from online sentiment data over the past four years, including market supply and demand, economic conditions, related agricultural prices, and market attention. Subsequently, using the SO-PMI algorithm to expand domain-specific lexicons and employing semantic and sentiment analysis with syntactic parsing, the study aims to improve the accuracy of sentiment quantification in text analysis. Furthermore, cointegration and Granger causality tests validate the significant impact of these factors on scallion price fluctuations, mitigating spurious regression issues. Finally, by employing a TVP-VAR model, the study compares the impulse responses and shock magnitudes of various factors in different time and spatial dimensions. It also delves into the mechanisms of heterogeneous impacts and trends in scallion prices, considering the corresponding socio-economic context and public sentiment events within specific time windows, thereby providing valuable decision-making insights for the healthy development of agricultural markets.

This study examined the eye movement patterns of 317 elementary students across reading conditions (audio-assisted reading (AR) and reading-only (R)) and text types (fiction and non-fiction), and identified eye movement parameters that predict their literal comprehension (LC) and inferential comprehension (IC). Participants, randomly assigned to either reading condition and either text type, answered questions assessing their LC and IC. Average fixation duration (AFD), total fixation duration (TFD), and scanpath length were used as eye movement parameters. There were main effects of age on all parameters and interaction effects between age and reading condition on TFD and scanpath length. Controlling for age, TFD had a positive impact on LC of both text types in the AR, while under the R, it affected negatively on IC of both text types. Longer scanpaths predicted IC of fiction in the AR, LC and IC of non-fiction under the AR, and LC of non-fiction within the R. AFD influenced negatively on IC of fiction in the AR, LC and IC of non-fiction in the AR, and LC of non-fiction under the R. These eye movement patterns and predictors indicate children employ different reading strategies based on reading modality and text type.

Background: Food insecurity is an ongoing problem in the U.S. with implications for health problems and social disadvantages. Past food insecurity intervention studies have targeted barriers to accessing healthy food. Although, information barriers have not been adequately addressed by food insecurity interventions. Methods: This qualitative study involved testing a text message as an informational nudge through interviews among adults in the southwest U.S. (n = 10). Interview questions focused on the message content, persuasiveness, relevancy, clarity, and details that participants liked and disliked. A qualitative content analysis of interview responses then highlighted any necessary modifications to the text message. Results: Interview responses highlighted that the message conveyed details about how to access food assistance resources. The message was perceived as informative, clear, and friendly. Participants highlighted the need for better spacing in the message layout, more details about the Supplemental Nutrition Assistance Program (SNAP), and why SNAP is relevant. Conclusion: Text message readability and informativeness were common concerns raised during interviews. Participant-informed modifications to the text message were applied in preparation for using the message in a food insecurity intervention. Future food insecurity research efforts should similarly test novel interventions to ensure the intervention is acceptable for the target population.

The emergence of the Covid-19 pandemic has created both negative and positive changes, including implementing the circular economy across the globe. This Systematic Review follows the PRISMA statement and employs the Text Mining (Voyant Tools) technique to visualize and analyze the impacts of the Covid-19 on three aspects of the circular economy: economic, social, and environmental. The research employs Latent Dirichlet Allocation (LDA) to identify five major topics: (1) Shortage of medical equipment but high medical waste during Covid-19 due to the high demand in healthcare; (2) The long term negative impacts of lockdown on economic and social activities because of Covid-19 pandemic; (3) The reports on impacts of Covid-19 pandemic on the manufacturing globally, and their coping strategies and new opportunities; (4) The impacts of international restriction on the tourism, trade, shipping, and aviation industry, causing billion-dollar losses; (5) The reduction of pollution with health environment improvements with example cases from China and EU. The research identifies current literature gaps in the circular economy and Covid- 19 topics and encourages the application of text mining tools into researching to stimulate the research process and assist in communicating with the public.

Google play store allow the user to download a mobile application (app) and user get inspired by the rating and reviews of the mobile app. A recent study analyzes that user preferences, user opinion for improvement, user sentiment about particular feature and detail with descriptions of experiences are very useful for an application developer. However, many application reviews are very large and difficult to process manually. Star rating is given of the whole application and the developer cannot analyze the single feature. In this research, we have scrapped 282,231 user reviews through different data scraping techniques. We have applied the text classification on these user reviews. We have applied different algorithms and find the precision, accuracy, F1 score and recall. In evaluated results, we have to also find the best algorithm.

Background. IgA nephropathy (IgAN) is associated with chronic inflammation. Platelet-related parameters, such as platelet (PLT) count, platelet-to-albumin ratio (PAR) and platelet-to-lymphocyte ratio (PLR), were examined as potential prognostic indicators for renal and cardiovascular (CV) outcomes in IgAN. Methods. 150 IgAN patients were devided into two groups based on the cut-off value of PAR. All-cause mortality, major CV events and end-stage renal disease were the primary combined endpoints. Secondary endpoints, such as CV or renal endpoints, were also analyzed separately. Results. The patient’s mean age was 43.7 ± 13.5 years, and the follow-up time was 124 ± 67 months. The K-M curve showed that PLR, PAR and PLT were strongly associated with primary combined (p = 0.002, p = 0.004, p = 0.001) and renal outcomes (p < 0.001, p < 0.001, p < 0.001), but not with CV outcomes in IgAN. However, when combined with left ventricular hypertrophy (LVH) or metabolic syndrome (MetS) PAR was found to be a significant predictor of both primary (p < 0.001, p < 0.001) and secondary outcomes (p = 0.001 and p = 0.038; p = 0.001 and p = 0.015). Additionally, PLR correlated with microalbuminuria (r = -0.165, p = 0.033) and LVH (r = -0.178, p = 0.025), while PLT correlated with eGFR (r = 0.158, p = 0.040). Conclusion. Elevated PAR and PLR may predict progression to end-stage kidney disease, but in combination with LVH and MetS, they showed CV events also in IgAN. The determination of PAR and PLR can be useful and cost-effective parameters for assessing both cardiovascular and renal risks in CKD.

Background: The aim of this study has been to investigate if inflammatory biomarkers such as the systemic immune-inflammation index (SII), neutrophil to lymphocyte ratio (NLR) and platelet to lymphocyte ratio (PLR) have any correlation with the recurrence of pleomorphic adenomas (PA) in patients afflicted by this disease. Methods: A retrospective multicentre study involving two Italian centres, the Maxillofacial Sur-gery at the Policlinico Federico II in Naples and the ENT and Maxillofacial Surgery at the IRCCS G. Pascale National Cancer Institute in Naples, has been performed. The study, involving 608 pa-tients who had received a diagnosis of PA, was conducted between January 2000 and January 2020. A Receiver Operating Characteristic (ROC) curve was used to calculate the optimal cut-off values for the SII, PLR and NLR. A linear regression model (LRM) and a non-linear logistic re-gression model (NLRM) were realized with respect to the NLR, PLR and SII. Results: The highest performance was achieved by the NLR (cut-off 2.960) with the AUC = 0.91, with an accuracy of 96.1%, sensitivity of 78.4% and specificity of 97.2%, PPV = 64.4% and NPV = 98.6%. The LRM and NLRM including all the haematological parameters did not increase the per-formance in terms of recurrence detection. Conclusion: The NLR is an excellent support tool in predicting the recurrence of PA with a cut-off of 2.960. It can be a good reference point for surgical decision-making and follow-up in clinical practice.

Background: Tuberculosis (TB) continues to be a serious public health threat that affects the most vulnerable population. Patients who are LTFU after TB diagnosis still represent one of the biggest challenges to TB control. Method: In this prospective observational study, we aimed to identify and analyse the risk factors associated with LTFU among TB patients who started first-line TB treatment in the Sanatorium Hospital in Luanda. Result: A total of 113 patients with TB were included between August 2018 and September 2019. Seventy-six (67.3%) patients were cured, 27 (23.9%) were LTFU, 5 (4.4%) died, 4 (3.5%) were transferred and 1 (0.9%) presented a treatment failure. After excluding those dead, transferred or who failed to treatment, we observed that severe TB at the time of diagnosis (OR 9.24, 95% CI 2.18-39.04) and food insecurity were significantly associated with LTFU (OR 5.96, 95% CI 1.66-21.41). Conclusion: The findings of our study are of enormous value to understand the reasons for the LTFU of patients with TB and can guide polices and facilitate designing measures to allow better adherence, and therefore, greater treatment success

Breast cancer (BC) affects millions of women; it is the leading cause of death by neoplasia in wom-en worldwide, with at least 500 thousand deaths each year, 70% of which occur in developing countries. Elastography evaluates tissue stiffness, making this a promising non-invasive real-time technique. BI-RADS is recognized as a standard and helpful technique for establishing BC risk. This study aimed to determine, by strain elastography (SE), the fat-to-lesion index (F/L index), sen-sitivity, specificity, and optimal cut-off to diagnose BC. This prospective study included 216 wom-en who underwent SE, ultrasound, mammography, and breast biopsy; according to mammogra-phy, we recruited BI-RADS category 2 (n= 5), category 3 (n=29), category 4 (n= 94), category 5 (n= 71) and category 6 (n=17); after biopsy, 108 women were classified as negative, and 108 were pos-itive for BC. Three expert BC radiologists performed all imaging interpretations and biopsies. Our results show mean values of the F/L index for negative biopsies of 3.70±2.57 and positive biopsies of 18.10±17.01; we developed two models for the prediction of BC, a logistic regression model with an AUC of 0.893 and a neural network with an AUC of 0.902. The Youden index yielded a F/L index cut-off >5.76, with a sensitivity and specificity of 84.26%. Positive Spearman's Rho cor-relation coefficient was observed between the F/L index and BI-RADS, r=.073 (p< 0.001). ANOVA test showed a difference between F/L index vs. molecular subtypes (p=0.002). SE is helpful as a di-agnostic complement to mammography, or in cases where the latter is contraindicated, elas-tography has an adequate predictive capacity and is a fast and minimally invasive procedure.

It has been proposed that upregulation of the noradrenergic (NA) system through exposure to novelty might be one of the key components for building cognitive reserve and resilience to neurodegeneration. A personality trait that promotes greater interest in novel experiences is Openness to Experience (OE). Individuals who are high in OE exhibit strong curiosity towards novelty and are more likely to be exposed to novel stimuli throughout the lifespan.Performing voxel-based morphometry analyses, we examined whether OE in 135 healthy young adults was associated with Locus Coeruleus (LC) - the main core of the noradrenergic system. We also investigated whether LC mediates the relationship between OE and a key measure of reserve, I.Q. In comparison with other neuromodulators’ seeds (Dorsal and Median Raphe [DR-MR] -5-HT, Ventral Tegmental Area [VTA] -DA- and Nucleus Basalis of Meynert [NBM] -Ach-), results show that greater LC integrity was related to greater OE and IQ. Additionally, mediation analyses revealed that only LC mediated the relationship between OE and IQ. Furthermore, a Bayesian multiple regression was conducted to model the contribution of both isolated and combined neuromodulator seeds to OE trait express. This approach found that the LC + VTA could account for more variance in OE expression compared to the LC alone. Other weaker but significant associations were observed between DR and OE, between LC + NBM and Conscientiousness and between DR + NBM and Extraversion. These results are together suggestive that OE is primarily underpinned by monoaminergic functioning, and that the LC-NA system plays a central role mediating the interdependence between OE expression and cognition (IQ level). While offering further understanding of the neurobiology of human personality, these findings indicate a pivotal role of LC-NA system integrity in novelty seeking in daily life. This provides a psychobiological account for how OE expression may be beneficial for maintaining the noradrenergic system to facilitate cognitive reserve and resilience to neurodegenerative conditions.

Background: Vaccinating children against COVID-19 is an essential public health strategy in order to reach herd immunity and prevent illness among children and adults. Parents are facing tremendous stress in relation to the COVID-19 pandemic and the effectiveness of the COVID-19 vaccination program for children. In this study, we aimed to investigate parents’ willingness to vaccinate their children against COVID-19 in North Kivu province (DRC). Methods: A cross-sectional survey between December 01, 2021 to January 20, 2022 in 6 health zones (Goma, Karisimbi, Butembo, Beni, Kamango and Katwa) was conducted in the province of North Kivu. In each health zone, we selected 5 clusters (Health area) using the method of probabilistic selection proportional to population size. In total, 522 parents participated in our study. Results: Overall, 32.8% of parents intended to vaccinate their children. In the multivariate analysis, younger age of parents (aOR : 2.40, CI : [1.50-3.83]), higher level of fear that “a member of my family” could contract COVID-19 (aOR : 2.35, CI : [1.38-4.02]), higher level of perceived vulnerability to COVID-19 within the family (aOR : 1.70, CI : [1.005-2.2881]), higher level of perceived susceptibility to COVID-19 within the family (aOR : 3.07, CI : [1.80-5.23]), and history of vaccination against COVID-19 among parents (aOR : 16.47, CI : [8.39-32.33]), were significantly associated with the intention to have their children or adolescents vaccinated. Conclusion: Willingness of parents to vaccinate their children against the COVID-19 vaccination was low in North Kivu. There is undeniably in this region a need to reinsure the populations about vaccine safety for both children and adults. Public health authorities should also address widespread misinformation about vaccines in a timely manner.

With the COVID-19 pandemic continuing and the resulting economic burden increasingly apparent, the Indonesian government began to prepare a "new normal" phase and make peace with COVID-19. From this new decision arises the question of the readiness of businesses and the industrial sector to resume operations amid COVID-19. This article aims to provide concise and precise information about the preparations that can be made by businesses to operate safely amid COVID-19 based on existing scientific studies and literatures. From the literature visits it can be concluded that transmission and danger of a COVID-19 pandemic can be prevented through: creation of infectious disease prevention and response plan, implementing basic infection prevention measures, policies and procedures for proper identification and isolation of sick people, applying flexibilities in policies, and protections in the workplace.

Here we review and describe a set of research priorities to meet present and future challenges posed to farmed animal production that build on progress, successes and resources from the Functional Annotation of ANimal Genomes (FAANG) project.

South America has become the new epicenter of the coronavirus, especially in Brazil where the disease continues to spread exponentially across the country. This text aims to analyze the psychosocial factors of Covid-19 on back to school strategies in Brazil from a bioecological perspective. At the microsystem level, the population is experiencing different levels of stress and fear; at the mesosystem level, changes in routines, separation from family and friends, and closure of schools; and at the macrosystem level, national guidelines to control the pandemic, institutional standards on a national and international scale. Therefore, the main focus for the success of school return must be in the prevention of contagion and with physical and psychological health, and should not only consider the demands of curricula, financial or administrative management. For this reason, it is paramount that greater female representativeness is increased in decision-making levels of the meso and macrosystem, regarding the resumption of school and academic activities in the pandemic period, since the number of female leaders in decision-making, is still insufficient.

The COVID-19 pandemic has presented the world with increased challenges. The response to this pandemic has led to a sudden disruption of routine medical and elective surgical care. Most hospitals across the globe have had to change the way outpatient clinics are carried out and postpone non-urgent elective surgical procedures. NHS England ceased all elective general surgeries to train and re-deploy their staff to support the increased pressures from COVID-19 in an intensive care setting. However, with a decline of reported cases and deaths, the return to undertaking non-urgent elective services is imminent. In May 2020, Radha and Afzal published the first evidence-based guidelines for the resumption of elective orthopaedic services titled “Evidence based suggestions for the return to elective orthopaedic surgery following the COVID-19 Pandemic”. The pathway presented is adapted from the “Evidence based suggestions for the return to elective orthopaedic surgery following the COVID-19 Pandemic” and could potentially be used as a model for other surgical specialities. We present a three-phased return back to urological services. Safe resumption of elective care is possible but needs to be carefully planned.

Most large fibers in central nervous system of vertebrates are decussated. The cause or the evolutionary force of this phenomenon is not convincingly described. Here we show a pattern for this nervous system property that simultaneously is found to a more basic rule in evolution.The idea here is the "function to anatomy" theorem. We claim that functions of physical, biochemical, enzymatic, etc., in lower scale of living organisms "emerge" to anatomy in larger scales. And reversely, the anatomy we see in multicelular organisms are/were functions in microscopic scales.We propose descriptions, predictions, and a probable road map to what this can implicate in biological and medical issues.

In 2014, the outbreak of adenoviral pneumonia occurred in Korean military training center. However, there is limited data on characteristics of fever and its response to antipyretics therapy in immunocompetent adults with adenoviral positive community-acquired pneumonia (CAP). Medical records of patients who were admitted to Armed Forces Chuncheon Hospital for treatment of CAP between January 2014 and December 2016 were retrospectively analyzed. We evaluated demographics, clinico-laboratory findings and radiologic findings at admission were compared between adenovirus positive (Adv) group and adenovirus negative (non-Adv) group. Out of 251 military personnel with CAP during the study periods, 67 were classified into Adv group while 184 were Non-Adv group. Patients with Adv group had a longer duration of fever after admission and symptom onset. Adv group patients had a higher mean temperature at admission and more observed over 40 and 39 to 40℃. Adv group patients had more commonly observed no response to antipyretic treatment and adverse events after antipyretics use. Length of hospital stay had no significant difference between two groups and no patient died in both groups. In our study, Adv positive CAP in patients with immunocompetent military personnel had distinct characteristics of fever and response to antipyretic treatment.

An understanding of the cultural conditions that determine the factors affecting nurses’ intention to leave is important for countries suffering from nurse shortage. Aim: to examine factors influencing intention to leave among female hospital nurses in a large Japanese sample, classified into four generations by age considering economic conditions. Methods: a cross-sectional survey with convenience sampling was conducted. Anonymous self-administered questionnaires were distributed to all nurses in 30 hospitals. To assess intention to leave, basic attributes, life conditions, work characteristics, and factors of psychosocial work environment were addressed. After classifying data into four generations based on age cohorts, we conducted multivariate logistic regression analysis using the completed data (N = 5,074, mean age = 36.24). Results: regardless of generational characteristics influenced by economic conditions, effort and monetary reward were generation-common factors. Over-commitment, social support, and the presence of a role model were generation-common factors in three generations. While having children increased intention to leave in the generation born 1965–1979, having family members in need of caregiving other than children decreased the risk in the generation born in the 1980s. Conclusion: generational countermeasures considering factors of psychosocial work environment and life conditions are needed to avert female nurse turnover.

Microaggressions are subtle verbal and non-verbal slights based on group membership, and they are ubiquitous in the lives of racial minorities, women, and LGBTQ individuals (Sue, 2010). The goal of the current paper is to introduce a role-playing based exercise on effective responses to microaggressions. The workshop draws on the prejudice responding workshops of Plous (2000) and Lawson, et al. (2010) but integrates research-based strategies. The activity was tested in two groups of undergraduate students, and the findings suggested that both groups felt more prepared to deal with microaggressions after participating.

Monochromatic documents claim for much less computer bandwidth for network transmission and storage space than their color or even grayscale equivalent. The binarization of historical documents is far more complex than recent ones as paper aging, color, texture, translucidity, stains, back-to-front interference, kind and color of ink used in handwritting, printing process, digitalization process, etc. are some of the factors that affect binarization. This article presents a new binarization algorithm for historical documents. The new global filter proposed is performed in four steps: filtering the image using a bilateral filter, splitting image into the RGB components, decision-making for each RGB channel based on an adaptive binarization method inspired by Otsu's method with a choice of the threshold level, and classification of the binarized images to decide which of the RGB components best preserved the document information in the foreground. The quantitative and qualitative assessment made with 21 binarization algorithms in three sets of “real world” documents showed very good results.

HER2/neu positive breast tumors predict a high mortality and comprise 25-30% of breast cancer. We have shown that Flavokawain A (FKA) preferentially reduces the viabilities of HER2 overexpressing breast cancer cell lines (i.e. SKBR3 and MCF7/HER2) versus those with less HER2 expression (i.e. MCF7 and MDA-MB-468). FKA at cytotoxic concentrations to breast cancer cell lines also has minimal effect on the growth of non-malignant breast epithelial MCF10 cells. FKA induces G2M arrest in cell cycle progression of HER2 overexpressing breast cancer cell lines through inhibition of Cdc2 and Cdc25C phosphorylation and down-regulation of Myt1 and Wee1expression leading to increased Cdc2 kinase activities. In addition, FKA induces apoptosis in SKBR3 cells by increasing the protein expression of Bim and BAX and decreasing expression of Bcl₂, Bcl_x/L , XIAP and survivin. FKA also down-regulates the protein expression of HER-2 and inhibits AKT phosphorylation. Herceptin plus FKA treatment leads to enhanced growth inhibitory effect on HER-2 overexpressing breast cancer cell lines through down-regulation of Myt1, Wee1, Skp2, Survivin and XIAP. Our results suggest the promise of FKA as a novel apoptosis inducer and G2 checkpoint abrogating agent in combination with Herceptin for treatment of HER2 overexpressing breast cancer.

Neutrophil to lymphocyte ratio (NLR) is a promising predictive and prognostic factor in breast cancer. We investigated its ability to predict disease-free survival (DFS) and overall survival (OS) in patients with luminal A or luminal B-HER2-negative breast cancer who received neoadjuvant chemotherapy (NACT). Pre-treatment complete blood cell counts from 168 consecutive patients with luminal breast cancer were evaluated to assess NLR. The study population was stratified into NLRlow or NLRhigh according to a cut-off value established by receiving operator curve (ROC) analysis. Data on additional pre- and post-treatment clinical-pathological characteristics were also collected. Kaplan-Meier curves, log-rank tests, and Cox proportional hazards models were used for statistical analyses. Patients with pre-treatment NLRlow showed a significantly shorter DFS (HR 6.97, 95% CI 1.65-10.55, p= 0.002) and OS (HR 7.79, 95% CI 1.25-15.07, p= 0.021) compared to those with NLRhigh. Non-ductal histology, luminal B subtype, and post-treatment Ki67≥ 14% were also associated with worse DFS (p= 0.016, p= 0.002, and p= 0.001, respectively). In multivariate analysis, luminal B subtype, post-treatment Ki67≥ 14%, and NLRlow remained independent prognostic factors for DFS, while only post-treatment Ki67≥ 14% and NLRlow affected OS. The present study provides evidence that pre-treatment NLRlow helps identify women at higher risk of recurrence and death among patients affected by luminal breast cancer treated with NACT.

Purpose: Hearing assistive technology (HAT) has been shown to be a viable solution to the speech-in-noise perception (SPIN) issue in children with autism spectrum disorder (ASD); however, little is known about its efficacy in tonal language speakers. This study compared sentence-level SPIN performance between Chinese children with ASD and neurotypical (NT) children and evaluated HAT use in improving SPIN performance and easing SPIN difficulty. Methods: Children with ASD (n=26) and NT children (n=19) aged 6-12 performed two adaptive tests in steady-state noise and three fixed-level tests in quiet and steady-state noise with and without using HAT. Speech recognition thresholds (SRT) and accuracy rates were assessed using adaptive and fixed-level tests, respectively. Parents or teachers of the ASD group completed a questionnaire regarding children’s listening difficulty under six circumstances before and after a ten-day trial period of HAT use. Results: Although the two groups of children had comparable SRTs, the ASD group showed a significantly lower SPIN accuracy rate than the NT group. Also, a significant impact of noise was found in the ASD group’s accuracy rate, but not in the NT group’s. There was a general improvement in the ASD group’s SPIN performance with HAT and a decrease in their listening difficulty ratings across all conditions after the device trial. Conclusion: The findings indicated inadequate SPIN in the ASD group using a relatively sensitive measure to gauge SPIN performance among children. The markedly increased accuracy rate in noise during HAT-on sessions for the ASD group confirmed the feasibility of HAT for improving SPIN performance in controlled laboratory settings, and the reduced post-use ratings of listening difficulty further confirmed the benefits of HAT use in daily scenarios.

The inaccurate estimates of linear prediction coefficient (LPC) and noise variance introduce bias in Kalman filter (KF) gain and degrades speech enhancement performance. The existing methods proposed a tuning of the biased Kalman gain particularly in stationary noise condition. This paper introduces a tuning of the KF gain for speech enhancement in real-life noise conditions. First, we estimate noise from each noisy speech frame using a speech presence probability (SPP) method to compute the noise variance. Then construct a whitening filter (with its coefficients computed from the estimated noise) and employed to the noisy speech, yielding a pre-whitened speech, from where the speech LPC parameters are computed. Then construct KF with the estimated parameters, where the robustness metric offsets the bias in Kalman gain during speech absence to that of the sensitivity metric during speech presence to achieve better noise reduction. Where the noise variance and the speech model parameters are adopted as a speech activity detector. The reduced-biased Kalman gain enables the KF to minimize the noise effect significantly, yielding the enhanced speech. Objective and subjective scores on NOIZEUS corpus demonstrates that the enhanced speech produced by the proposed method exhibits higher quality and intelligibility than some benchmark methods.

The emergence of Artificial Intelligence (AI) driven Keyword Spotting (KWS) technologies has revolutionized human to machine interaction. Yet, the challenge of end-to-end energy efficiency, memory footprint and system complexity of current Neural Network (NN) powered AI-KWS pipelines has remained ever present. This paper evaluates KWS utilizing a learning automata powered machine learning algorithm called the Tsetlin Machine (TM). Through significant reduction in parameter requirements and choosing logic over arithmetic based processing, the TM offers new opportunities for low-power KWS while maintaining high learning efficacy. In this paper we explore a TM based keyword spotting (KWS) pipeline to demonstrate low complexity with faster rate of convergence compared to NNs. Further, we investigate the scalability with increasing keywords and explore the potential for enabling low-power on-chip KWS.

Text classification is one of the most important task in natural language processing, In this research, we carried out several experimental research on three (3) of the most popular Text classification NLP classifier in Convolutional Neural Network (CNN), Multinomial Naive Bayes (MNB), and Support Vector Machine (SVN). In the presence of enough training data, Deep Learning CNN works best in all parameters for evaluation with 77% accuracy, followed by SVM with accuracy of 76%, and multinomial Bayes with least performance of 69% accuracy. CNN has the best performance in the presence of large enough training dataset because of the presence of filter/ kernels which help to indentify patterns in text data regardless of their position in the sentence. We repeated the training again with just one-third of our data, at this point SVM comes with the best performance, the performance of CNN noticeably drops but still better than Multinomial Naive Bayes, the reason why SVM works best when we reduce the training data was because of its ability to look for a hyper-plane that creates a boundary between different classes of data so as to properly classify them, so we believed that getting the hyper-plane was more efficient when we reduce the dataset, hence reason for the good performance. Multinomial Naive Bayes have the least performance which we attributed to its assumption of independency between the features which sometimes does not hold true. We concluded that availability of data should be an important factor when choosing classifier for Natural Language Processing Text Classification task. CNN should be use in the presence of enough dataset, and SVM should be use when data is not enough. Multinomial Naive Bayes must not be trusted with state of the art NLP task due to its assumption of independency between the features.

Tourism is an important industry that promotes national economic and social progress. All-for-one tourism is a new concept of regionally coordinated development that uses the tourism industry as an engine to boost resource integration, industrial integration, and social sharing. Tourism service is the main embodiment of cultural ecosystem services for all-for-one tourism cities. Taking Liyang in China as an example, this paper used a combination of GIS spatial analysis and big data text mining to evaluate the tourism service from three aspects: the quality of tourism resources, the comprehensiveness of tourism service facilities, and the satisfaction of tourists. The results show that (1) the tourism service is better in the northwestern and eastern areas of the city, while it is lower in the northeastern and southwestern parts; (2) the hotspot areas should focus on improving the tour routes, the transport capacity, and excessive charges; the cold spot areas should work on ecological restoration and creating new tourism attractions by combining the local industries; (3) rural tourism integrating agriculture and visitation should be highlighted as a key growth point to improve the city’s tourism service function.

Wide-scale commercialization of cultured meat, produced from in vitro-grown animal stem cells, is still a long way off because of technical, regulatory and, above all, social acceptability constraints. As a result, despite the advancement of knowledge, it is difficult to predict whether this innovation will be commonly accepted. The concept of media framing is key to better understanding individuals’ decision-making and mental interpretative patterns in relation to cultured meat adoption. This research therefore analyzes the role of media framing (namely the social media platform Twitter) on the components of consumer attitudes (cognitive, affective, and conative) in relation to cultured meat. A total of 23 020 publications and 38 531 comments were analyzed qualitatively (content analysis) and quantitatively (MANOVA). This study showed that media-framed posts influenced consumer attitudes more than did non-media-framed posts. Although the results indicate that the different types of media framing (ethical, intrinsic, informational, and belief) do not exert the same influence on each attitude component, they suggest that posts combining the ethical, intrinsic, and informational media frames have a greater impact on the acceptability of cultured meat and that the belief frame is equally important, particularly for the behavioral component. Relevant implications can be drawn for authorities and businesses on the use of differentiated education and marketing strategies

The use of artificial intelligence in natural language processing (NLP) has significantly contributed to the advancement of natural language applications such as sentimental analysis, topic modeling, text classification, chatbots, and spam filtering. With a large amount of text generated each day from different sources such as webpages, blogs, emails, social media, and articles, one of the most common tasks in natural language processing is the classification of a text corpus. This is important in many institutions for planning, decision-making, and archives of their projects. Many algorithms exist to automate text classification operations but the most intriguing of them is that which also learns these operations automatically. In this study, we present a new model to infer and learn from data using probabilistic logic and apply it to text classification. This model, called GenCo, is a multi-input single-output (MISO) learning model that uses a collaboration of partial classifications to generate the desired output. It provides a heterogeneity measure to explain its classification results and enables the reduction of the curse of dimensionality in text classification. The classification results are compared with those of conventional text classification models, and it shows that our proposed model has a higher classification performance than conventional models.

In today's business world, many companies have realized that consumer feedback evaluations play a crucial part in shaping the company's future activities. The hotel and tourism industries are well-known examples where the user feedback has become crucial. Even though most accommodations want their clients to express their opinions about their services, these reviews are nevertheless carefully reviewed by them. This manual approach is time demanding and error prone, the level of expertise of the individual analyzing the evaluations determines the quality of the manual analysis. This research focuses on the development of a tool-based support system that analyzes text-based customer evaluations and then highlights the maintenance issues identified in the reviews, removing the need to manually evaluate and analyze the customer reviews. The process includes data preparation, sentiment analysis, and a fully trained deep learning model to extract critical insights. Using sentiment analysis techniques after data preprocessing and trained deep learning model, an accuracy of 96% was achieved, and this paper discusses how this technology helps to meet the needs of accommodation management by completing activities efficiently. This might save management time and helps to save money without losing clients, allowing them to enhance their additional sales.

In today's business world, many companies have realized that consumer feedback evaluations play a crucial part in shaping the company's future activities. The hotel and tourism industries are well-known examples where the user feedback has become crucial. Even though most accommodations want their clients to express their opinions about their services, these reviews are nevertheless carefully reviewed by them. This manual approach is time demanding and error prone, the level of expertise of the individual analyzing the evaluations determines the quality of the manual analysis. This research focuses on the development of a tool-based support system that analyzes text-based customer evaluations and then highlights the maintenance issues identified in the reviews, removing the need to manually evaluate and analyze the customer reviews. The process includes data preparation, sentiment analysis, and a fully trained deep learning model to extract critical insights. Using sentiment analysis techniques after data preprocessing and trained deep learning model, an accuracy of 96\% was achieved, and this paper discusses how this technology helps to meet the needs of accommodation management by completing activities efficiently. This might save management time and helps to save money without losing clients, allowing them to enhance their additional sales.

Background: HIV poses a threat to global health. With effective treatment options available, education and testing strategies are essential in preventing transmission. Text messaging is an effective tool for health promotion and can be used to target higher risk populations. This study reports on the design, delivery and testing of a mobile text messaging SMS intervention for HIV prevention and awareness, aimed at adults in the construction industry and delivered during the COVID-19 pandemic. Method: Participants were recruited at Test@Work workplace health promotion events (21 sites, n=464 employees), including health checks with HIV testing. Message development was based on a participatory design and included a focus group (n=9) and message fidelity testing (n=291) with assessment of intervention uptake, reach, acceptability, and engagement. Barriers to HIV testing were identified and mapped to the COM-B behavioural model. 23 one-way push SMS messages (19 included short web links) were generated and fidelity tested, then sent via automated SMS to two employee cohorts over a 10-week period during the COVID-19 pandemic. Engagement metrics measured were; opt-outs, SMS delivered/read, number of clicks per web link, and four two-way pull messages exploring repeat HIV testing, learning new information, perceived usefulness and behaviour change. Results: 291 people participated (68.3% of eligible attendees). A total of 7,726 messages were sent between March and June 2020, with 91.6% successfully delivered (100% read). 12.4% of participants opted out over 10 weeks. Of delivered messages, links were clicked an average of 14.4%, max 24.1% for HIV related links. The number of clicks on web links declined over time (r= -6.24, p=0.01). Response rate for two-way pull messages was 13.7% of participants. Since the workplace HIV test offer at recruitment, 21.6% reported having taken a further HIV test. Qualitative replies indicated behavioural influence of messaging on exercise, lifestyle behaviours and intention to HIV test. Conclusion: SMS messaging for HIV prevention and awareness is acceptable to adults in the construction industry, has high uptake, low attrition and good engagement with message content, when delivered during a global pandemic. Data collection methods may need refinement for audience and effect of COVID-19 on results is yet to be understood.