Information, Volume 15, Issue 6 (June 2024) – 35 articles

In the digital age, the intersection of artificial intelligence (AI) and higher education (HE) poses novel ethical considerations, necessitating a comprehensive exploration of this multifaceted relationship. This study aims to quantify and characterize the current research trends and critically assess the discourse on ethical AI applications within HE. Employing a mixed-methods design, we integrated quantitative data from the Web of Science, Scopus, and the Lens databases with qualitative insights from selected studies to perform scientometric and content analyses, yielding a nuanced landscape of AI utilization in HE. Our results identified vital research areas through citation bursts, keyword co-occurrence, and thematic clusters. We provided a conceptual model for ethical AI integration in HE, encapsulating dichotomous perspectives on AI’s role in education. Three thematic clusters were identified: ethical frameworks and policy development, academic integrity and content creation, and student interaction with AI. The study concludes that, while AI offers substantial benefits for educational advancement, it also brings challenges that necessitate vigilant governance to uphold academic integrity and ethical standards. The implications extend to policymakers, educators, and AI developers, highlighting the need for ethical guidelines, AI literacy, and human-centered AI tools. Full article

Enhancing the livelihood environment for fishermen’s communities with the rapid technological growth is essential in the marine sector. Among the various issues in the fishing industry, fishing zone identification and fish catch detection play a significant role in the fishing community. In this work, the automated prediction of potential fishing zones and classification of fish species in an aquatic environment through machine learning algorithms is developed and implemented. A prototype of the boat structure is designed and developed with lightweight wooden material encompassing all necessary sensors and cameras. The functions of the unmanned boat (FishID-AUV) are based on the user’s control through a user-friendly mobile/web application (APP). The different features impacting the identification of hotspots are considered, and feature selection is performed using various classifier-based learning algorithms, namely, Naive Bayes, Nearest neighbors, Random Forest and Support Vector Machine (SVM). The performance of classifications are compared. From the real-time results, it is clear that the Naive Bayes classification model is found to provide better accuracy, which is employed in the application platform for predicting the potential fishing zone. After identifying the first catch, the species are classified using an AlexNet-based deep Convolutional Neural Network. Also, the user can fetch real-time information such as the status of fishing through live video streaming to determine the quality and quantity of fish along with information like pH, temperature and humidity. The proposed work is implemented in a real-time boat structure prototype and is validated with data from sensors and satellites. Full article

The growing number of vehicles on the roads has resulted in several challenges, including increased accident rates, fuel consumption, pollution, travel time, and driving stress. However, recent advancements in intelligent vehicle technologies, such as sensors and communication networks, have the potential to revolutionize road traffic and address these challenges. In particular, the concept of platooning for autonomous vehicles, where they travel in groups at high speeds with minimal distances between them, has been proposed to enhance the efficiency of road traffic. To achieve this, it is essential to determine the precise position of vehicles relative to each other. Global positioning system (GPS) devices have an intended positioning error that might increase due to various conditions, e.g., the number of available satellites, nearby buildings, trees, driving into tunnels, etc., making it difficult to compute the exact relative position between two vehicles. To address this challenge, this paper proposes a new architectural framework to improve positioning accuracy using images captured by onboard cameras. It presents a novel algorithm and performance results for vehicle positioning based on GPS and video data. This approach is decentralized, meaning that each vehicle has its own camera and computing unit and communicates with nearby vehicles. Full article

With the rapid growth of the data scale in data centers, the high reliability of storage is facing various challenges. Specifically, hardware failures such as disk faults occur frequently, causing serious system availability issues. In this context, hardware fault prediction based on AI and big data technologies has become a research hotspot, aiming to guide operation and maintenance personnel to implement preventive replacement through accurate prediction to reduce hardware failure rates. However, existing methods still have weaknesses in terms of accuracy due to the impacts of data quality issues such as the sample imbalance. This article proposes a disk fault prediction method based on classification intensity resampling, which fills the gap between the degree of data imbalance and the actual classification intensity of the task by introducing a base classifier to calculate the classification intensity, thus better preserving the data features of the original dataset. In addition, using ensemble learning methods such as random forests, combined with resampling, an integrated classifier for imbalanced data is developed to further improve the prediction accuracy. Experimental verification shows that compared with traditional methods, the F1-score of disk fault prediction is improved by 6%, and the model training time is also greatly reduced. The fault prediction method proposed in this paper has been applied to approximately 80 disk drives and nearly 40,000 disks in the production environment of a large bank’s data center to guide preventive replacements. Compared to traditional methods, the number of preventive replacements based on our method has decreased by approximately 21%, while the overall disk failure rate remains unchanged, thus demonstrating the effectiveness of our method. Full article

To avoid the overfitting problem of the network model and improve the facial expression recognition effect of partially occluded facial images, an improved facial expression recognition algorithm based on MobileViT has been proposed. Firstly, in order to obtain features that are useful and richer for experiments, deep convolution operations are added to the inverted residual blocks of this network, thus improving the facial expression recognition rate. Then, in the process of dimension reduction, the activation function can significantly improve the convergence speed of the model, and then quickly reduce the loss error in the training process, as well as to preserve the effective facial expression features as much as possible and reduce the overfitting problem. Experimental results on RaFD, FER2013, and FER2013Plus show that this method has significant advantages over mainstream networks and the network achieves the highest recognition rate. Full article

In the era of Internet of Things (IoT) we have entered, the “Monitoring–Decision–Execution” cycle of typical autonomic and automation systems is extended, so it includes distributed developments that might scale from a smart home or greenhouse to a smart city and from autonomous driving to emergency management [...] Full article

This paper provides empirical evidence using two studies to explain the primary factors facilitating electronic health record (EHR) systems adoption through the lens of the resource advantage theory. We aim to address the following research questions: What are the main organizational antecedents of EHR implementation? What is the role of monitoring in EHR system implementation? What are the current themes and people’s attitudes toward EHR systems? This paper includes two empirical studies. Study 1 presents a research model based on data collected from four different archival datasets. Drawing upon the resource advantage theory, this paper uses archival data from 200 Texas hospitals, thus mitigating potential response bias and enhancing the validity of the findings. Study 2 includes a text analysis of 5154 textual data, sentiment analysis, and topic modeling. Study 1’s findings reveal that joint ventures and ownership are the two main enablers of adopting EHR systems in 200 Texas hospitals. Moreover, the results offer a moderating role of monitoring in strengthening the relationship between joint-venture capability and the implementation of EHR systems. Study 2’s results indicate a positive attitude toward EHR systems. The U.S. was unique in the sample due to its slower adoption of EHR systems than other developed countries. Physician burnout also emerged as a significant concern in the context of EHR adoption. Topic modeling identified three themes: training, healthcare interoperability, and organizational barriers. In a multimethod design, this paper contributes to prior work by offering two new EHR antecedents: hospital ownership and joint-venture capability. Moreover, this paper suggests that the monitoring mechanism moderates the adoption of EHR systems in Texas hospitals. Moreover, this paper contributes to prior EHR works by performing text analysis of textual data to carry out sentiment analysis and topic modeling. Full article

Finding passages related to a sentence over a large collection of text documents is a fundamental task for claim verification and open-domain question answering. For instance, a common approach for verifying a claim is to extract short snippets of relevant text from a collection of reference documents and provide them as input to a natural language inference machine that determines whether the claim can be deduced or refuted. Available approaches struggle when several pieces of evidence from different documents need to be combined to make an inference, as individual documents often have a low relevance with the input and are therefore excluded. We propose GRAAL (GRAph-based retrievAL), a novel graph-based approach that outlines the relevant evidence as a subgraph of a large graph that summarizes the whole corpus. We assess the validity of this approach by building a large graph that represents co-occurring entity mentions on a corpus of Wikipedia pages and using this graph to identify candidate text relevant to a claim across multiple pages. Our experiments on a subset of FEVER, a popular benchmark, show that the proposed approach is effective in identifying short passages related to a claim from multiple documents. Full article

Fog computing (FC) is a distributed architecture in which computing resources and services are placed on edge devices closer to data sources. This enables more efficient data processing, shorter latency times, and better performance. Fog computing was shown to be a promising solution for addressing the new computing requirements. However, there are still many challenges to overcome to utilize this new computing paradigm, in particular, reliability and security. Following this need, a systematic literature review was conducted to create a list of requirements. As a result, the following four key requirements were formulated: (1) low latency and response times; (2) scalability and resource management; (3) fault tolerance and redundancy; and (4) privacy and security. Low delay and response can be achieved through edge caching, edge real-time analyses and decision making, and mobile edge computing. Scalability and resource management can be enabled by edge federation, virtualization and containerization, and edge resource discovery and orchestration. Fault tolerance and redundancy can be enabled by backup and recovery mechanisms, data replication strategies, and disaster recovery plans, with a residual number system (RNS) being a promising solution. Data security and data privacy are manifested in strong authentication and authorization mechanisms, access control and authorization management, with fully homomorphic encryption (FHE) and the secret sharing system (SSS) being of particular interest. Full article

Recently, the widespread use of social media and easy access to the Internet have brought about a significant transformation in the type of textual data available on the Web. This change is particularly evident in Arabic language usage, as the growing number of users from diverse domains has led to a considerable influx of Arabic text in various dialects, each characterized by differences in morphology, syntax, vocabulary, and pronunciation. Consequently, researchers in language recognition and natural language processing have become increasingly interested in identifying Arabic dialects. Numerous methods have been proposed to recognize this informal data, owing to its crucial implications for several applications, such as sentiment analysis, topic modeling, text summarization, and machine translation. However, Arabic dialect identification is a significant challenge due to the vast diversity of the Arabic language in its dialects. This study introduces a novel hybrid machine and deep learning model, incorporating an attention mechanism for detecting and classifying Arabic dialects. Several experiments were conducted using a novel dataset that collected information from user-generated comments from Twitter of Arabic dialects, namely, Egyptian, Gulf, Jordanian, and Yemeni, to evaluate the effectiveness of the proposed model. The dataset comprises 34,905 rows extracted from Twitter, representing an unbalanced data distribution. The data annotation was performed by native speakers proficient in each dialect. The results demonstrate that the proposed model outperforms the performance of long short-term memory, bidirectional long short-term memory, and logistic regression models in dialect classification using different word representations as follows: term frequency-inverse document frequency, Word2Vec, and global vector for word representation. Full article

Digital technologies occupy an important place in today’s developing world. They are also strongly related to new trends in educational technologies. In this context, the digital accessibility of this new environment for people with various special needs is of particular concern. A novel tool for assessment of the technological ecosystem, designed to provide digital accessibility to people with special needs, is described in the paper. The overall structure and the initial test of the system are discussed in the paper. The conceptual framework of the ecosystem and its ontological model are described. Special attention is paid to the accessibility of digital learning and e-learning for people with special needs from a robotic perspective. Full article

Recent research emphasizes the importance of Artificial Intelligence applications as supporting tools for students in higher education. Simultaneously, an intensive exchange of views has started in the public debate in the international educational community. However, for a more proper use of these applications, it is necessary to investigate the factors that explain their intention and actual use in the future. With the Unified Theory of Acceptance and Use of Technology (UTAUT2) model, this work analyses the factors influencing students’ use and intention to use Artificial Intelligence technology. For this purpose, a sample of 197 Greek students at the School of Humanities and Social Sciences from the University of Patras participated in a survey. The findings highlight that expected performance, habit, and enjoyment of these Artificial Intelligence applications are key determinants influencing teachers’ intentions to use them. Moreover, behavioural intention, habit, and facilitating conditions explain the usage of these Artificial Intelligence applications. This study did not reveal any moderating effects. The limitations, practical implications, and proposed directions for future research based on these results are discussed. Full article

Personalized recommender systems play a crucial role in assisting users in discovering items of interest from vast amounts of information across various domains. However, developing accurate personalized recommender systems remains challenging due to the need to balance model architectures, input feature combinations, and fusion of heterogeneous data sources. This study investigates the impacts of these factors on recommendation performance using the MovieLens and Book Recommendation datasets. Six models, including single-task neural networks, multi-task learning, and baselines, were evaluated with various input feature combinations using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The multi-task learning approach achieved significantly lower RMSE and MAE by effectively leveraging heterogeneous data sources for personalized recommendations through a shared neural network architecture. Furthermore, incorporating user data and content data progressively enhanced performance compared to using only item identifiers. The findings highlight the importance of advanced model architectures and fusing heterogeneous data sources for high-quality recommendations, providing valuable insights for designing effective recommender systems across diverse domains. Full article

The evolving role of artificial intelligence (AI) in healthcare can shift the route of automated, supervised and computer-aided diagnostic radiology. An extensive literature review was conducted to consider the potential of designing a fully automated, complete diagnostic platform capable of integrating the current medical imaging technologies. Adjuvant, targeted, non-systematic research was regarded as necessary, especially to the end-user medical expert, for the completeness, understanding and terminological clarity of this discussion article that focuses on giving a representative and inclusive idea of the evolutional strides that have taken place, not including an AI architecture technical evaluation. Recent developments in AI applications for assessing various organ systems, as well as enhancing oncology and histopathology, show significant impact on medical practice. Published research outcomes of AI picture segmentation and classification algorithms exhibit promising accuracy, sensitivity and specificity. Progress in this field has led to the introduction of the concept of explainable AI, which ensures transparency of deep learning architectures, enabling human involvement in clinical decision making, especially in critical healthcare scenarios. Structure and language standardization of medical reports, along with interdisciplinary collaboration between medical and technical experts, are crucial for research coordination. Patient personal data should always be handled with confidentiality and dignity, while ensuring legality in the attribution of responsibility, particularly in view of machines lacking empathy and self-awareness. The results of our literature research demonstrate the strong potential of utilizing AI architectures, mainly convolutional neural networks, in medical imaging diagnostics, even though a complete automated diagnostic platform, enabling full body scanning, has not yet been presented. Full article

One of the practical obstacles limiting the use of cloud-based gamification applications is the lack of an Internet connection of adequate quality. In this paper, we describe a practical solution to this problem by the implementation of client-side gamification rule processing so that most events generated by players can be processed without the need to involve server-side functions; therefore, only a handful of data have to be transmitted to the server for global state synchronization, and only when an Internet connection is available. For this purpose, we adopt a simple textual gamification rule definition format, implement the rule parser and event processor, and evaluate the solution in terms of performance in experimental conditions. The obtained results are optimistic, showing that the developed solution can easily handle rule sets and event streams of realistic sizes. The solution is planned to be integrated into the next version of the FGPE gamified programming education platform. Full article

With the rapid development of new power systems, diverse new power services have imposed stricter requirements on network resources and performance. However, the traditional method of transmitting request data to the IoT management platform for unified processing suffers from large delays due to long transmission distances, making it difficult to meet the delay requirements of new power services. Therefore, to reduce the transmission delay, data transmission, storage and computation need to be performed locally. However, due to the limited resources of individual nodes in the local power wireless communication network, issues such as tight coupling between devices and resources and a lack of flexible allocation need to be addressed. The collaborative allocation of resources among multiple nodes in the local network is necessary to satisfy the multi-dimensional resource requirements of new power services. In response to the problems of limited node resources, inflexible resource allocation, and the high complexity of multi-dimensional resource allocation in local power wireless communication networks, this paper proposes a multi-objective joint optimization model for the collaborative allocation of communication, storage, and computing resources. This model utilizes the computational characteristics of communication resources to reduce the dimensionality of the objective function. Furthermore, a mouse swarm optimization algorithm based on multi-strategy improvements is proposed. The simulation results demonstrate that this method can effectively reduce the total system delay and improve the utilization of network resources. Full article

In the age of digitalization, business-to-business (B2B) data sharing is becoming increasingly important, enabling organizations to collaborate and make informed decisions as well as simplifying operations and hopefully creating a cost-effective virtual value chain. This is crucial to the success of modern businesses, especially global business. However, this approach also comes with significant privacy and security challenges, thus requiring robust mechanisms to protect sensitive information. After analyzing the evolving status of B2B data sharing, the purpose of this study is to provide insights into the design of theoretical framework solutions for the field. This study adopts technologies including encryption, access control, data anonymization, and audit trails, with the common goal of striking a balance between facilitating data sharing and protecting data confidentiality as well as data integrity. In addition, emerging technologies such as homomorphic encryption, blockchain, and their applicability as well as advantages in the B2B data sharing environment are explored. The results of this study offer a new approach to managing complex data sharing between organizations, providing a strategic mix of traditional and innovative solutions to promote secure and efficient digital collaboration. Full article

Large language models (LLMs) have the ability to generate text by stringing together words from their extensive training data. The leading AI text generation tool built on LLMs, ChatGPT, has quickly grown a vast user base since its release, but the domains in which it is being heavily leveraged are not yet known to the public. To understand how generative AI is reshaping print media and the extent to which it is being implemented already, methods to distinguish human-generated text from that generated by AI are required. Since college students have been early adopters of ChatGPT, we sought to study the presence of generative AI in newspaper articles written by collegiate journalists. To achieve this objective, an accurate AI detection model is needed. Herein, we analyzed university newspaper articles from different universities to determine whether ChatGPT was used to write or edit the news articles. We developed a detection model using classical machine learning and used the model to detect AI usage in the news articles. The detection model showcased a 93% accuracy in the training data and had a similar performance in the test set, demonstrating effectiveness in AI detection above existing state-of-the-art detection tools. Finally, the model was applied to the task of searching for generative AI usage in 2023, and we found that ChatGPT was not used to revise articles to any appreciable measure to write university news articles at the schools we studied. Full article

The management of water resources is becoming increasingly important in several contexts, including agriculture. Recently, innovative agricultural practices, advanced sensors, and Internet of Things (IoT) devices have made it possible to improve the efficiency of water use. However, it is the application of control strategies based on advanced machine learning techniques that enables the adoption of smart irrigation scheduling and the immediate economic, social, and environmental benefits. This challenging research area has attracted the attention of many researchers worldwide, who have proposed several technological and methodological solutions. Unfortunately, the results of these scientific efforts have not yet been categorized in a thematic survey, making it difficult to understand how far we are from optimal water management based on machine learning. This paper fills this gap by focusing on smart irrigation systems with an emphasis on machine learning. More specifically, the generic structure of a smart agriculture system is presented, and existing machine learning strategies and available datasets are discussed. Furthermore, several open issues are identified, especially in the processing of long-term data, also due to the lack of corresponding annotated datasets. Finally, some interesting future research directions to be pursued in order to build scalable, domain-independent approaches are proposed. Full article

Forming optimal peer groups is a crucial step in multiplier valuation. Among others, the traditional regression methodology requires the definition of the optimal set of peer selection criteria and the optimal size of the peer group a priori. Since there exists no universally applicable set of closed and complementary rules on selection criteria due to the complexity and the diverse nature of firms, this research exclusively examines unlisted companies, rendering direct comparisons with existing studies impractical. To address this, we developed a bespoke benchmark model through rigorous regression analysis. Our aim was to juxtapose its outcomes with our unique approach, enriching the understanding of unlisted company transaction dynamics. To stretch the performance of the linear regression method to the maximum, various datasets on selection criteria (full as well as F- and NCA-optimized) were employed. Using a sample of over 20,000 private firm transactions, model performance was evaluated employing multiplier prediction error measures (emphasizing bias and accuracy) as well as prediction superiority directly. Emphasizing five enterprise and equity value multiples, the results allow for the overall conclusion that the self-organizing map algorithm outperforms the traditional linear regression model in both minimizing the valuation error as measured by the multiplier prediction error measures as well as in direct prediction superiority. Consequently, the machine learning methodology offers a promising way to improve peer selection in private firm multiplier valuation. Full article

The advent of Internet of Things (IoT) devices has revolutionized our daily routines, fostering interconnectedness and convenience. However, this interconnected network also presents significant security challenges concerning authentication and data integrity. Traditional security measures, such as Public Key Infrastructure (PKI), encounter limitations when applied to resource-constrained IoT devices. This paper proposes a novel decentralized PKI system tailored specifically for IoT environments to address these challenges. Our approach introduces a unique “zone” architecture overseen by zone masters, facilitating efficient certificate management within IoT clusters while reducing the risk of single points of failure. Furthermore, we prioritize the use of lightweight cryptographic techniques, including Elliptic Curve Cryptography (ECC), to optimize performance without compromising security. Through comprehensive evaluation and benchmarking, we demonstrate the effectiveness of our proposed solution in bolstering the security and efficiency of IoT ecosystems. This contribution underlines the critical need for innovative security solutions in IoT deployments and presents a scalable framework to meet the evolving demands of IoT environments. Full article

With the development of the Internet, the oversight of research integrity issues has extended beyond the scientific community to encompass the whole of society. If these issues are not addressed promptly, they can significantly impact the research credibility of both institutions and scholars. This article proposes a text convolutional neural network based on SMOTE to identify short texts of potential public opinion events related to suspected scientific integrity issues from common short texts. The SMOTE comprehensive sampling technique is employed to handle imbalanced datasets. To mitigate the impact of short text length on text representation quality, the Doc2vec embedding model is utilized to represent short text, yielding a one-dimensional dense vector. Additionally, the dimensions of the input layer and convolution kernel of TextCNN are adjusted. Subsequently, a short text event extraction model based on TF-IDF and TextRank is proposed to extract crucial information, for instance, names and research-related institutions, from events and facilitate the identification of potential public opinion events related to suspected scientific integrity issues. Results of experiments have demonstrated that utilizing SMOTE to balance the dataset is able to improve the classification results of TextCNN classifiers. Compared to traditional classifiers, TextCNN exhibits greater robustness in addressing the problems of imbalanced datasets. However, challenges such as low information content, non-standard writing, and polysemy in short texts may impact the accuracy of event extraction. The framework can be further optimized to address these issues in the future. Full article

Stemming vulnerabilities out of a smart contract prior to its deployment is essential to ensure the security of decentralized applications. As such, numerous tools and machine-learning-based methods have been proposed to help detect vulnerabilities in smart contracts. Furthermore, various ways of encoding the smart contracts for analysis have also been proposed. However, the impact of these input methods has not been systematically studied, which is the primary goal of this paper. In this preliminary study, we experimented with four common types of input, including Word2Vec, FastText, Bag-of-Words (BoW), and Term Frequency–Inverse Document Frequency (TF-IDF). To focus on the comparison of these input types, we used the same deep-learning model, i.e., convolutional neural networks, in all experiments. Using a public dataset, we compared the vulnerability detection performance of the four input types both in the binary classification scenarios and the multiclass classification scenario. Our findings show that TF-IDF is the best overall input type among the four. TF-IDF has excellent detection performance in all scenarios: (1) it has the best F1 score and accuracy in binary classifications for all vulnerability types except for the delegate vulnerability where TF-IDF comes in a close second, and (2) it comes in a very close second behind BoW (within 0.8%) in the multiclass classification. Full article

Electroencephalography (EEG), electromyography (EMG), galvanic skin response (GSR), and electrocardiogram (ECG) are among the techniques developed for collecting psychophysiological data from humans. This study presents a feature extraction technique for identifying emotions in EEG-based data from the human brain. Independent component analysis (ICA) was employed to eliminate artifacts from the raw brain signals before applying signal extraction to a convolutional neural network (CNN) for emotion identification. These features were then learned by the proposed CNN-LSTM (long short-term memory) algorithm, which includes a ResNet-152 classifier. The CNN-LSTM with ResNet-152 algorithm was used for the accurate detection and analysis of human emotional data. The SEED V dataset was employed for data collection in this study, and the implementation was carried out using an Altera DE2 FPGA development board, demonstrating improved performance in terms of FPGA speed and area optimization. Full article

Thanks to the publication of ChatGPT, Artificial Intelligence is now basically accessible and usable to all internet users. The technology behind it can be used in many chatbots, whereby the chatbots should be trained for the respective area of application. Depending on the application, the chatbot should react differently and thus, for example, also take on and embody personality traits to be able to help and answer people better and more personally. This raises the question of whether ChatGPT-4 is able to embody personality traits. Our study investigated whether ChatGPT-4’s personality can be analyzed using personality tests for humans. To test possible approaches to measuring the personality traits of ChatGPT-4, experiments were conducted with two of the most well-known personality tests: the Big Five and Myers–Briggs. The experiments also examine whether and how personality can be changed by user input and what influence this has on the results of the personality tests. Full article

This study delves into the dual nature of artificial intelligence (AI), illuminating its transformative potential that has the power to revolutionize various aspects of our lives. We delve into critical issues such as AI hallucinations, misinformation, and unpredictable behavior, particularly in large language models (LLMs) and AI-powered chatbots. These technologies, while capable of manipulating human decisions and exploiting cognitive vulnerabilities, also hold the key to unlocking unprecedented opportunities for innovation and progress. Our research underscores the need for robust, ethical AI development and deployment frameworks, advocating a balance between technological advancement and societal values. We emphasize the importance of collaboration among researchers, developers, policymakers, and end users to steer AI development toward maximizing benefits while minimizing potential harms. This study highlights the critical role of responsible AI practices, including regular training, engagement, and the sharing of experiences among AI users, to mitigate risks and develop the best practices. We call for updated legal and regulatory frameworks to keep pace with AI advancements and ensure their alignment with ethical principles and societal values. By fostering open dialog, sharing knowledge, and prioritizing ethical considerations, we can harness AI’s transformative potential to drive human advancement while managing its inherent risks and challenges. Full article

Recognizing fraudulent activity in the banking system is essential due to the significant risks involved. When fraudulent transactions are vastly outnumbered by non-fraudulent ones, dealing with imbalanced datasets can be difficult. This study aims to determine the best model for detecting fraud by comparing four commonly used machine learning algorithms: Support Vector Machine (SVM), XGBoost, Decision Tree, and Logistic Regression. Additionally, we utilized the Synthetic Minority Over-sampling Technique (SMOTE) to address the issue of class imbalance. The XGBoost Classifier proved to be the most successful model for fraud detection, with an accuracy of 99.88%. We utilized SHAP and LIME analyses to provide greater clarity into the decision-making process of the XGBoost model and improve overall comprehension. This research shows that the XGBoost Classifier is highly effective in detecting banking fraud on imbalanced datasets, with an impressive accuracy score. The interpretability of the XGBoost Classifier model was further enhanced by applying SHAP and LIME analysis, which shed light on the significant features that contribute to fraud detection. The insights and findings presented here are valuable contributions to the ongoing efforts aimed at developing effective fraud detection systems for the banking industry. Full article

In recent times, the prevalence of chatbot technology has notably increased, particularly in the realm of medical assistants. However, there is a noticeable absence of medical chatbots that cater to the Hungarian language. Consequently, Hungarian-speaking people currently lack access to an automated system capable of providing assistance with their health-related inquiries or issues. Our research aims to establish a competent medical chatbot assistant that is accessible through both a website and a mobile app. It is crucial to highlight that the project’s objective extends beyond mere linguistic localization; our goal is to develop an official and effectively functioning Hungarian chatbot. The assistant’s task is to answer medical questions, provide health advice, and inform users about health problems and treatments. The chatbot should be able to recognize and interpret user-provided text input and offer accurate and relevant responses using specific algorithms. In our work, we put a lot of emphasis on having steady input so that it can detect all the diseases that the patient is dealing with. Our database consisted of sentences and phrases that a user would type into a chatbot. We assigned health problems to these and then assigned the categories to the corresponding cure. Within the research, we developed a website and mobile app, so that users can easily use the assistant. The app plays a particularly important role for users because it allows them to use the assistant anytime and anywhere, taking advantage of the portability of mobile devices. At the current stage of our research, the precision and validation accuracy of the system is greater than 90%, according to the selected test methods. Full article

Rationale. Object tracking has significance in many applications ranging from control of unmanned vehicles to autonomous monitoring of specific situations and events, especially when providing safety for patients with certain adverse conditions such as epileptic seizures. Conventional tracking methods face many challenges, such as the need for dedicated attached devices or tags, influence by high image noise, complex object movements, and intensive computational requirements. We have developed earlier computationally efficient algorithms for global optical flow reconstruction of group velocities that provide means for convulsive seizure detection and have potential applications in fall and apnea detection. Here, we address the challenge of using the same calculated group velocities for object tracking in parallel. Methods. We propose a novel optical flow-based method for object tracking. It utilizes real-time image sequences from the camera and directly reconstructs global motion-group parameters of the content. These parameters can steer a rectangular region of interest surrounding the moving object to follow the target. The method successfully applies to multi-spectral data, further improving its effectiveness. Besides serving as a modular extension to clinical alerting applications, the novel technique, compared with other available approaches, may provide real-time computational advantages as well as improved stability to noisy inputs. Results. Experimental results on simulated tests and complex real-world data demonstrate the method’s capabilities. The proposed optical flow reconstruction can provide accurate, robust, and faster results compared to current state-of-the-art approaches. Full article