https://doi.org/10.5220/0013818300004067

Mental health challenges among PhD scholars are a growing global concern, with a survey in the UK revealing that at least 86% of students face depression and anxiety. Social media platforms offer valuable insights into the depression levels of PhD students. Sentiment analysis for social media content can help identify indicators of anxiety, such as negative language, stress expressions, or mental health struggles. This paper uses social media and surveys to develop a dataset for Pakistani graduate students. The dataset collects 5096 social media posts from 1170 users, categorising them into anxiety (46.7%), depression (12.6%), and motivation (40.7%) based on mental health levels. The survey responses are combined with the social media dataset. Machine learning models, including Support Vector Machine (SVM), Artificial Neural Network (ANN), and Random Forest (RF), are used to detect the mental health status of PhD scholars. The study finds that 59.3% of graduate students in Pakistan face anxiety and mental health issues, indicating a need for policy reformulation in graduate programmes. The research data is available online for further research (https://github.com/dr-m-wasim/PhD-Scholars-Mental-Health).

Abstract Lexical answer type prediction is integral to biomedical question–answering systems. LAT prediction aims to predict the expected answer’s semantic type of a factoid or list-type biomedical question. It also aids in the answer processing stage of a QA system to assign a high score to the most relevant answers. Although considerable research efforts exist for LAT prediction in diverse domains, it remains a challenging biomedical problem. LAT prediction for the biomedical field is a multi-label classification problem, as one biomedical question might have more than one expected answer type. Achieving high performance on this task is challenging as biomedical questions have limited lexical features. One biomedical question must be assigned multiple labels given these limited lexical features. In this paper, we develop a novel feature set (lexical, noun concepts, verb concepts, protein–protein interactions, and biomedical entities) from these lexical features. Using ensemble learning with bagging, we use the label power set transformation technique to classify multi-label. We evaluate the integrity of our proposed methodology on the publicly available multi-label biomedical questions dataset (MLBioMedLAT) and compare it with twelve state-of-the-art multi-label classification algorithms. Our proposed method attains a micro-F1 score of 77%, outperforming the baseline model by 25.5%.

The manufacturing industry relies on continuous optimization to meet quality and safety standards, which is part of the Industry 4.0 concept.Predicting when a specific part of a product will fail to meet these standards is of utmost importance and requires vast amounts of data, which often are collected from variety of sensors, often reffered to as Industrial Internet of Things (IIoT).Using a published dataset from Bosch, that describes the process at every step of production, we aim to train a machine learning model that can accurately predict faults in the manufacturing process.The dataset provides two years of production data across four production lines and 52 stations.Considering that the data generated from each production part includes more than four thousand features, we investigate various feature selection and data preprocessing methods.The obtained results exhibit Area Under the Receiver Operating Characteristic Curve (AUC ROC) of up to 0.997, which is remarkable and promising even for real-life production use.

The search for a reliable cloud service provider has become increasingly complex due to the increasing reliance on them. Researchers have developed a multi-criteria decision-making (MCDM) method, including the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) for ranking. However, the selection process often relies on only a few aspects of service quality, known as Quality of Service (QoS) parameters. This research proposes a comprehensive methodology, using the Consensus Assessment Initiative Questionnaire (CAIQ) to select standardized QoS parameters and then applying the MCDM method, incorporating TOPSIS, to rank cloud service providers based on their performance. This approach aims to improve the accuracy and reliability of the selection process, enabling individuals and organizations to confidently identify and choose the most trustworthy cloud service providers for their specific needs.

Viruses significantly threaten computer systems, potentially causing extensive damage and data loss. All users must prioritize cybersecurity by installing effective antivirus software, safeguarding their PCs against potential harm. Even though there are many different kinds of malware, ransomware is particularly dangerous since it prevents victims from accessing their vital data or locks files permanently unless they pay a ransom to the attackers. Recent ransomware strains must be categorized promptly. Data for the present investigation was gathered from a variety of web resources, including Kaggle and ransomware.re. Concerning using Kaggle to acquire harmless datasets, ransomware.re is retrieved for use in a study on ransomware. Many preprocessing methods, such as Normalisation and Imputation, are used to polish our datasets. The most recent additions to the dataset were classified using the Random Forest tree classifier, with a final accuracy of 99.9%. Random Forest Tree fared exceptionally well compared to the KNN and SVM algorithms. We also highlighted that additional preprocessing methods can enhance outcomes for SVM and KNN.

The use of the big data in conjunction with artificial intelligence methods used in health areas is increasingly being used. For data capture, smartphones and embedded sensors are an increasingly reliable, accurate and fast way to detect certain types of bio signals. In this study we propose a system for collecting data for the acquisition and study of associated diseases and symptoms to demonstrate how the use of sensors and their connection to a database and later application of methods to detect patterns to remove conclusions and contribute to advances in health and to a higher quality of life especially in the elderly.

Stress is or could be one of the most talked-about and recurring things in recent years, because of the world that we live. Stress is our body's response to a pressure thing or situation in our life. In this way, countless things have a stressful impact on our lives. On the other hand, stress is usually due to something new or unexpected, that somehow is beyond our control. The effects on our body are evident, inevitably having symptoms. When we are exposed to stress, certain hormones are released in our body, and the immune system is working on self-defence. During this, breathing becomes faster, heart rate increases, muscles contract and blood pressure also increases. Thus, the organism is ready to act to protect itself. It is where our project comes in, because, with these symptoms of our body, they allow stress to be identified. This paper is focused on precisely that, because, by reading the person's vital data, we can establish standards of normality, which, when they suffer variation, may indicate to us in advance that the person is stressed and help him to control himself so that there is no more significant damage. With this, we hope to obtain positive results in people's lives and routine, causing the stress rate to drop worldwide.

Generally, a massive number of mobile applications is growing day today. There are several types of applications. However, healthcare applications are critical domain nowadays in the scientific field. Consequently, it is crucial to understand how the current state of the art in this domain in non-European countries is such as Georgia. Therefore, this paper presents a study concerning the current scenario on e-health applications which is available for Georgian citizens. Furthermore, this paper examines which are the barriers of development of e-health and m-health in the before mentioned country. The results show a limited number of existing applications. This study analysis 11 mobile applications. In total, 55% of the analyzed apps only support the Georgian language and 36% are informative mobile applications. However, the available mobile apps in the Georgian language does not provide communication between doctors and patients which is a critical limitation. The results will be helpful for further developments of mobile apps in this field in Georgia.

The functional tests are essential to test the functionality of different types of people, and specialty for older adults. The primary purpose of this paper is to create a method for the automatic measurement of the results of the different functional tests. These are the Heel-rise Test, Functional Reach Test, Timed Up and Go Test, Ten Meter Walk Test, Eight hop test, Up-down hop test, Side hop test, Single hop test, Chair Stand Test, Arm Curl Test, and Chair Sit and Reach test. The use of sensors may increase the accuracy of the measurements of these tests. These tests may identify several diseases, and it will be subject to further research in the future.

The recognition of Activities of Daily Living (ADL) and their environments based on sensors available in off-the-shelf mobile devices is an emerging topic. These devices are capable to acquire and process the sensors' data for the correct recognition of the ADL and their environments, providing a fast and reliable feedback to the user. However, the methods implemented in a mobile application for this purpose should be adapted to the low resources of these devices. This paper focuses on the demonstration of a mobile application that implements a framework, that forks their implementation in several modules, including data acquisition, data processing, data fusion and classification methods based on the sensors? data acquired from the accelerometer, gyroscope, magnetometer, microphone and Global Positioning System (GPS) receiver. The framework presented is a function of the number of sensors available in the mobile devices and implements the classification with Deep Neural Networks (DNN) that reports an accuracy between 58.02% and 89.15%.