Information, Volume 15, Issue 6 (June 2024) – 62 articles

Multi-Access Edge Computing (MEC) reduces latency, provides high-bandwidth applications with real-time performance and reliability, supporting new applications and services for the present and future Beyond the Fifth Generation (B5G). Radio Network Information Service (RNIS) plays a crucial role in obtaining information from the Radio Access Network (RAN). With the advent of 5G, RNIS requires improvements to handle information from the new generations of RAN. In this scenario, improving the RNIS is essential to boost new applications according to the strict requirements imposed. Hence, this work proposes a new RNIS as a service to the MEC framework in B5G networks to improve MEC applications. The service is validated and evaluated, and demonstrates the ability to adequately serve a large number of MEC apps (two, four, six and eight) and from 100 to 2000 types of User Equipment (UE). Full article

In the field of visualization, understanding users’ analytical reasoning is important for evaluating the effectiveness of visualization applications. Several studies have been conducted to capture and analyze user interactions to comprehend this reasoning process. However, few have successfully linked these interactions to users’ reasoning processes. This paper introduces an approach that addresses the limitation by correlating semantic user interactions with analysis decisions using an interactive wire transaction analysis system and a visual state transition matrix, both designed as visual analytics applications. The system enables interactive analysis for evaluating financial fraud in wire transactions. It also allows mapping captured user interactions and analytical decisions back onto the visualization to reveal their decision differences. The visual state transition matrix further aids in understanding users’ analytical flows, revealing their decision-making processes. Classification machine learning algorithms are applied to evaluate the effectiveness of our approach in understanding users’ analytical reasoning process by connecting the captured semantic user interactions to their decisions (i.e., suspicious, not suspicious, and inconclusive) on wire transactions. With the algorithms, an average of 72% accuracy is determined to classify the semantic user interactions. For classifying individual decisions, the average accuracy is 70%. Notably, the accuracy for classifying ‘inconclusive’ decisions is 83%. Overall, the proposed approach improves the understanding of users’ analytical decisions and provides a robust method for evaluating user interactions in visualization tools. Full article

Channel estimation accuracy significantly affects the performance of orthogonal frequency-division multiplexing (OFDM) systems. In the literature, there are quite a few channel estimation methods. However, the performances of these methods deteriorate considerably when the wireless channels suffer from nonlinear distortions and interferences. Machine learning (ML) shows great potential for solving nonparametric problems. This paper proposes ML-based channel estimation methods for systems with comb-type pilot patterns and random pilot symbols, such as ATSC 3.0. We compare their performances with conventional channel estimations in ATSC 3.0 systems for linear and nonlinear channel models. We also evaluate the robustness of the ML-based methods against channel model mismatch and signal-to-noise ratio (SNR) mismatch. The results show that the ML-based channel estimations achieve good mean squared error (MSE) performance for linear and nonlinear channels if the channel statistics used for the training stage match those of the deployment stage. Otherwise, the ML estimation models may overfit the training channel, leading to poor deployment performance. Furthermore, the deep neural network (DNN)-based method does not outperform the linear channel estimation methods in nonlinear channels. Full article

The design of automotive human–machine interfaces (HMIs) for global consumers’ needs to cater to a broad spectrum of drivers. This paper comprises benchmark studies and explores how users from international markets—Germany, China, and the United States—engage with the same automotive HMI. In real driving scenarios, N = 301 participants (premium vehicle owners) completed several tasks using different interaction modalities. The multi-method approach included both self-report measures to assess preference and satisfaction through well-established questionnaires and observational measures, namely experimenter ratings, to capture interaction performance. We observed a trend towards lower preference ratings in the Chinese sample. Further, interaction performance differed across the user groups, with self-reported preference not consistently aligning with observed performance. This dissociation accentuates the importance of integrating both measures in user studies. By employing benchmark data, we provide insights into varied market-based perspectives on automotive HMIs. The findings highlight the necessity for a nuanced approach to HMI design that considers diverse user preferences and interaction patterns. Full article

SimRank is a widely used metric for evaluating vertex similarity based on graph topology, with diverse applications such as large-scale graph mining and natural language processing. The objective of the single-source and top-k SimRank query problem is to retrieve the kvertices with the largest SimRank to the source vertex. However, existing algorithms suffer from inefficiency as they require computing SimRank for all vertices to retrieve the top-k results. To address this issue, we propose an algorithm named HitSimthat utilizes a branch and bound strategy for the single-source and top-k query. HitSim initially partitions vertices into distinct sets based on their shortest-meeting lengths to the source vertex. Subsequently, it computes an upper bound of SimRank for each set. If the upper bound of a set is no larger than the minimum value of the current top-k results, HitSim efficiently batch-prunes the unpromising vertices within the set. However, in scenarios where the graph becomes dense, certain sets with large upper bounds may contain numerous vertices with small SimRank, leading to redundant overhead when processing these vertices. To address this issue, we propose an optimized algorithm named HitSim-OPT that computes the upper bound of SimRank for each vertex instead of each set, resulting in a fine-grained and efficient pruning process. The experimental results conducted on six real-world datasets demonstrate the performance of our algorithms in efficiently addressing the single-source and top-k query problem. Full article

In addressing the nuanced interplay between consumer attitudes and Artificial Intelligence (AI) use readiness in physical retail stores, the main objective of this study is to test the impacts of prior experience, as well as perceived risks with AI technologies, self-assessment of consumers’ ability to manage AI technologies, and the moderator role of gender in this relationship. Using a quantitative cross-sectional survey, data from 243 consumers familiar with AI technologies were analyzed using structural equation modeling (SEM) methods to explore these dynamics in the context of physical retail stores. Additionally, the moderating impacts were tested after the invariance analysis across both gender groups. Key findings indicate that positive prior experience with AI technologies positively influences AI use readiness in physical retail stores, while perceived risks with AI technologies serve as a deterrent. Gender differences significantly moderate these effects, with perceived risks with AI technologies more negatively impacting women’s AI use readiness and self-assessment of the ability to manage AI technologies showing a stronger positive impact on men’s AI use readiness. The study concludes that retailers must consider these gender-specific perceptions and attitudes toward AI to develop more effective strategies for technology integration. Our research also highlights the need to address gender-specific barriers and biases when adopting AI technology. Full article

The use of large language models (LLMs) is now spreading in several areas of research and development. This work is concerned with systematically reviewing LLMs’ involvement in engineering education. Starting from a general research question, two queries were used to select 370 papers from the literature. Filtering them through several inclusion/exclusion criteria led to the selection of 20 papers. These were investigated based on eight dimensions to identify areas of engineering disciplines that involve LLMs, where they are most present, how this involvement takes place, and which LLM-based tools are used, if any. Addressing these key issues allowed three more specific research questions to be answered, offering a clear overview of the current involvement of LLMs in engineering education. The research outcomes provide insights into the potential and challenges of LLMs in transforming engineering education, contributing to its responsible and effective future implementation. This review’s outcomes could help address the best ways to involve LLMs in engineering education activities and measure their effectiveness as time progresses. For this reason, this study addresses suggestions on how to improve activities in engineering education. The systematic review on which this research is based conforms to the rules of the current literature regarding inclusion/exclusion criteria and quality assessments in order to make the results as objective as possible and easily replicable. Full article

The preservation of historical traditional architectural ensembles faces multifaceted challenges, and the need for facade renovation and updates has become increasingly prominent. In conventional architectural updating and renovation processes, assessing design schemes and the redesigning component are often time-consuming and labor-intensive. The knowledge-driven method utilizes a wide range of knowledge resources, such as historical documents, architectural drawings, and photographs, commonly used to guide and optimize the conservation, restoration, and management of architectural heritage. Recently, the emergence of artificial intelligence-generated content (AIGC) technologies has provided new solutions for creating architectural facades, introducing a new research paradigm to the renovation plans for historic districts with their variety of options and high efficiency. In this study, we propose a workflow combining Grasshopper with Stable Diffusion: starting with Grasshopper to generate concise line drawings, then using the ControlNet and low-rank adaptation (LoRA) models to produce images of traditional Minnan architectural facades, allowing designers to quickly preview and modify the facade designs during the renovation of traditional architectural clusters. Our research results demonstrate Stable Diffusion’s precise understanding and execution ability concerning architectural facade elements, capable of generating regional traditional architectural facades that meet architects’ requirements for style, size, and form based on existing images and prompt descriptions, revealing the immense potential for application in the renovation of traditional architectural groups and historic districts. It should be noted that the correlation between specific architectural images and proprietary term prompts still requires further addition due to the limitations of the database. Although the model generally performs well when trained on traditional Chinese ancient buildings, the accuracy and clarity of more complex decorative parts still need enhancement, necessitating further exploration of solutions for handling facade details in the future. Full article

Recently, the research community has shown significant interest in the continuous temporal data obtained from motion sensors in wearable devices. These data are useful for classifying and analysing different human activities in many application areas such as healthcare, sports and surveillance. The literature has presented a multitude of deep learning models that aim to derive a suitable feature representation from temporal sensory input. However, the presence of a substantial quantity of annotated training data is crucial to adequately train the deep networks. Nevertheless, the data originating from the wearable devices are vast but ineffective due to a lack of labels which hinders our ability to train the models with optimal efficiency. This phenomenon leads to the model experiencing overfitting. The contribution of the proposed research is twofold: firstly, it involves a systematic evaluation of fifteen different augmentation strategies to solve the inadequacy problem of labeled data which plays a critical role in the classification tasks. Secondly, it introduces an automatic feature-learning technique proposing a Multi-Branch Hybrid Conv-LSTM network to classify human activities of daily living using multimodal data of different wearable smart devices. The objective of this study is to introduce an ensemble deep model that effectively captures intricate patterns and interdependencies within temporal data. The term “ensemble model” pertains to fusion of distinct deep models, with the objective of leveraging their own strengths and capabilities to develop a solution that is more robust and efficient. A comprehensive assessment of ensemble models is conducted using data-augmentation techniques on two prominent benchmark datasets: CogAge and UniMiB-SHAR. The proposed network employs a range of data-augmentation methods to improve the accuracy of atomic and composite activities. This results in a 5% increase in accuracy for composite activities and a 30% increase for atomic activities. Full article

Cybersecurity is a crucial concern for local governments as they serve as the primary interface between public and government services, managing sensitive data and critical infrastructure. While technical safeguards are integral to cybersecurity, the role of a well-structured policy is equally important as it provides structured guidance to translate technical requirements into actionable protocols. This study reviews local governments’ cybersecurity policies to provide a comprehensive assessment of how these policies align with the National Institute of Standards and Technology’s Cybersecurity Framework 2.0, which is a widely adopted and commonly used cybersecurity assessment framework. This review offers local governments a mirror to reflect on their cybersecurity stance, identifying potential vulnerabilities and areas needing urgent attention. This study further extends the development of a cybersecurity policy framework, which local governments can use as a strategic tool. It provides valuable information on crucial cybersecurity elements that local governments must incorporate into their policies to protect confidential data and critical infrastructure. Full article

As autonomous driving technology advances, the imperative of ensuring pedestrian traffic safety becomes increasingly prominent within the design framework of autonomous driving systems. Pedestrian trajectory prediction stands out as a pivotal technology aiming to address this challenge by striving to precisely forecast pedestrians’ future trajectories, thereby enabling autonomous driving systems to execute timely and accurate decisions. However, the prevailing state-of-the-art models often rely on intricate structures and a substantial number of parameters, posing challenges in meeting the imperative demand for lightweight models within autonomous driving systems. To address these challenges, we introduce Social Spatio-Temporal Graph Multi-Layer Perceptron (Social-STGMLP), a novel approach that utilizes solely fully connected layers and layer normalization. Social-STGMLP operates by abstracting pedestrian trajectories into a spatio-temporal graph, facilitating the modeling of both the spatial social interaction among pedestrians and the temporal motion tendency inherent to pedestrians themselves. Our evaluation of Social-STGMLP reveals its superiority over the reference method, as evidenced by experimental results indicating reductions of 5% in average displacement error (ADE) and 17% in final displacement error (FDE). Full article

Within the literary domain, genres function as fundamental organizing concepts that provide readers, publishers, and academics with a unified framework. Genres are discrete categories that are distinguished by common stylistic, thematic, and structural components. They facilitate the categorization process and improve our understanding of a wide range of literary expressions. In this paper, we introduce a new dataset for genre classification of Russian books, covering 11 literary genres. We also perform dataset evaluation for the tasks of binary and multi-class genre identification. Through extensive experimentation and analysis, we explore the effectiveness of different text representations, including stylometric features, in genre classification. Our findings clarify the challenges present in classifying Russian literature by genre, revealing insights into the performance of different models across various genres. Furthermore, we address several research questions regarding the difficulty of multi-class classification compared to binary classification, and the impact of stylometric features on classification accuracy. Full article

Light-field images (LFIs) are gaining increased attention within the field of 3D imaging, virtual reality, and digital refocusing, owing to their wealth of spatial and angular information. The escalating volume of LFI data poses challenges in terms of storage and transmission. To address this problem, this paper introduces an MSHPE (most-similar hierarchical prediction encoding) structure based on light-field multi-view images. By systematically exploring the similarities among sub-views, our structure obtains residual views through the subtraction of the encoded view from its corresponding reference view. Regarding the encoding process, this paper implements a new encoding scheme to process all residual views, achieving lossless compression. High-efficiency video coding (HEVC) is applied to encode select residual views, thereby achieving lossy compression. Furthermore, the introduced structure is conceptualized as a layered coding scheme, enabling progressive transmission and showing good random access performance. Experimental results demonstrate the superior compression performance attained by encoding residual views according to the proposed structure, outperforming alternative structures. Notably, when HEVC is employed for encoding residual views, significant bit savings are observed compared to the direct encoding of original views. The final restored view presents better detail quality, reinforcing the effectiveness of this approach. Full article

In human resource management, training is considered one of the most effective ways to improve employees’ task performance. However, the effectiveness of training depends mostly on the resources and effort of learners, especially the operant resources. This study investigates the influence of operant resources on individual task performance within the framework of knowledge management. Building on existing research, a quantitative model was developed and tested using data from 296 Vietnamese managers and senior employees. Data analysis employed SPSS 21 and AMOS 24 software. The findings provide strong support for all nine proposed hypotheses, demonstrating a positive impact of operant resources on both learner behavior and subsequent task performance. The research highlights the significant role of individual operant resources in enhancing learning outcomes and employee effectiveness. Managerial implications are derived from these results, offering practical guidance for businesses to improve training activities and ultimately boost employee task performance. Full article

The emerging digitalization in today’s industrial environments allows manufacturers to store online knowledge about production and use it to make better informed management decisions. This paper proposes a multi-agent framework enhanced with digital twin (DT) for production scheduling and optimization. Decentralized scheduling agents interact to efficiently manage the work allocation in different segments of production. A DT is used to evaluate the performance of different scheduling decisions and to avoid potential risks and bottlenecks. Production managers can supervise the system’s decision-making processes and manually regulate them online. The multi-agent system (MAS) uses asset administration shells (AASs) for data modelling and communication, enabling interoperability and scalability. The framework was deployed and tested in an industrial pilot coming from the bicycle production industry, optimizing and controlling the short-term production schedule of the different departments. The evaluation resulted in a higher production rate, thus achieving higher production volume in a shorter time span. Managers were also able to coordinate schedules from different departments in a dynamic way and achieve early bottleneck detection. Full article

Imitation learning has shown promise for autonomous driving, but suffers from covariate shift, where the policy performs poorly in unseen environments. DAgger is a popular approach that addresses this by leveraging expert demonstrations. However, DAgger’s frequent visits to sub-optimal states can lead to several challenges. This paper proposes a novel DAgger framework that integrates Bayesian uncertainty estimation via mean field variational inference (MFVI) to address this issue. MFVI provides better-calibrated uncertainty estimates compared to prior methods. During training, the framework identifies both uncertain and critical states, querying the expert only for these states. This targeted data collection reduces the burden on the expert and improves data efficiency. Evaluations on the CARLA simulator demonstrate that our approach outperforms existing methods, highlighting the effectiveness of Bayesian uncertainty estimation and targeted data aggregation for imitation learning in autonomous driving. Full article

Workload management is a cornerstone of contemporary human resource management with widespread applications in private and public sectors. The challenges in human resource management are particularly pronounced within the public sector: particularly in task allocation. The absence of a standardized workload distribution method presents a significant challenge and results in unnecessary costs in terms of man-hours and financial resources expended on surplus human resource utilization. In the current research, we analyze how to deal with the “race condition” above and propose a dynamic workload management model based on the response time required to implement each task. Our model is trained and tested using comprehensive employee data comprising 450 records for training, 100 records for testing, and 88 records for validation. Approximately 11% of the initial data are deemed either inaccurate or invalid. The deployment of the ANFIS algorithm provides a quantified capability for each employee to handle tasks in the public sector. The proposed idea is deployed in a virtualized platform where each employee is implemented as an independent node with specific capabilities. An upper limit of work acceptance is proposed based on a documented study and laws that suggest work time frames in each public body, ensuring that no employee reaches the saturation level of exhaustion. In addition, a variant of the “slow start” model is incorporated as a hybrid congestion control mechanism with exceptional outcomes, offering a gradual execution window for each node under test and providing a smooth and controlled start-up phase for new connections. The ultimate goal is to identify and outline the entire structure of the Greek public sector along with the capabilities of its employees, thereby determining the organization’s executive capacity. Full article

This study conducts a bibliometric analysis of online customer and hotel review research, aiming to provide insights into where each field comes from, stands now and ought to go in the future. In particular, this study examines how the existing research on online customer reviews can benefit future hotel review research. Data collected from Web-of-Science and Scopus created an expanded network of 797 core articles and 19,374 citations to identify intellectual structures, developing trends, and future research gaps. This study offers a visual overview of journals, institutions, countries, research themes and authors to assess the overall directions hotels can take. It underscores the necessity for rigorous and relevant research amid the proliferation of online reviews and emphasises the imperative for academia to bridge the gap between theoretical insights and practical applications within the dynamic tourism industry. This study provides researchers and industry professionals with useful tools to understand and deal with the complexities of online reviews. It also highlights the important role these reviews play in shaping the future of tourism strategies. Full article

Sustainable mobility is one of the main challenges on a global level. In this context, the emerging Mobility as a Service (MaaS) plays an important role in the mobility of people. This paper investigates the main enabling factors for implementing the MaaS paradigm, with a specific focus on the level of acceptance of this new technology. To achieve this objective, the proposed methodology for measuring the potential MaaS acceptance is based on a set of pilot surveys. The methodology integrates motivational surveys with Stated and Revealed Preference (SP, RP) and Technology Acceptance Models (TAM). The collected data are processed to obtain indicators that measure the potential level of MaaS acceptance. The main results of the two pilot experiments are illustrated by referring to urban and extra-urban mobility with or without physical barriers. The results obtained show that the level of MaaS acceptance grows with the increase in generalized transport costs perceived by the users. Full article

Text matching, as a core technology of natural language processing, plays a key role in tasks such as question-and-answer systems and information retrieval. In recent years, the development of neural networks, attention mechanisms, and large-scale language models has significantly contributed to the advancement of text-matching technology. However, the rapid development of the field also poses challenges in fully understanding the overall impact of these technological improvements. This paper aims to provide a concise, yet in-depth, overview of the field of text matching, sorting out the main ideas, problems, and solutions for text-matching methods based on statistical methods and neural networks, as well as delving into matching methods based on large-scale language models, and discussing the related configurations, API applications, datasets, and evaluation methods. In addition, this paper outlines the applications and classifications of text matching in specific domains and discusses the current open problems that are being faced and future research directions, to provide useful references for further developments in the field. Full article

Diagnostic classification models (DCM) are latent structure models with discrete multivariate latent variables. Recently, extensions of DCMs to mixed membership have been proposed. In this article, ordinary DCMs, mixed and partial membership models, and multidimensional item response theory (IRT) models are compared through analytical derivations, three example datasets, and a simulation study. It is concluded that partial membership DCMs are similar, if not structurally equivalent, to sufficiently complex multidimensional IRT models. Full article

The research and management of Industry 4.0 increasingly relies on accurate real-time quality data to apply efficient algorithms for predictive maintenance. Currently, Low-Power Wide-Area Networks (LPWANs) offer potential advantages in monitoring tasks for predictive maintenance. However, their applicability requires improvements in aspects such as energy consumption, transmission range, data rate and constant quality of service. Commonly used battery-operated IIoT devices have several limitations in their adoption in large facilities or heat-intensive industries (iron and steel, cement, etc.). In these cases, the self-heating nodes together with the appropriate low-power processing platform and industrial sensors are aligned with the requirements and real-time criteria required for industrial monitoring. From an environmental point of view, the carbon footprint associated with human activity leads to a steady rise in global average temperature. Most of the gases emitted into the atmosphere are due to these heat-intensive industries. In fact, much of the energy consumed by industries is dissipated in the form of waste heat. With this scenario, it makes sense to build heat transformation collection systems as guarantors of battery-free self-powered IIoT devices. Thermal energy harvesters work on the physical basis of the Seebeck effect. In this way, this paper gathers the methodology that standardizes the modelling and simulation of waste heat recovery systems for IoT nodes, gathering energy from any hot surface, such as a pipe or chimney. The statistical analysis is carried out with the data obtained from two different IoT architectures showing a good correlation between model simulation and prototype behaviour. Additionally, the selected model will be coupled to a low-power processing platform with LoRaWAN connectivity to demonstrate its effectiveness and self-powering ability in a real industrial environment. Full article

This study aims to examine the influence of social customer relationship management (CRM) on relationship quality (RQ); the role of strategic information exchange capabilities (SIECs) as a mediator on the relationship between dimensions of social CRM and RQ was also investigated. A self-structured questionnaire survey was conducted on the subordinates working at various family-style restaurants in Egypt. Following a simple random sampling procedure, 466 valid responses were used for data analysis. The findings reveal that three dimensions of social CRM, namely customer service quality (CSQ), integrated marketing channels (IMCs), and online communities (OCs), have statistically significant effects on RQ. Moreover, SIECs mediate the relationship between CS/IMCs/OCs and RQ. The other two dimensions, rewards (RDs) and value-added services (VSs), do not directly or indirectly affect RQ. This study opens new avenues in the existing literature by identifying the most relevant factors affecting RQ in the context of Egyptian restaurants. This study can enable policymakers and restaurant owners to formulate social CRM strategies and achieve customer satisfaction properly. This study explores the mediation mechanism of SIECs on the relationship between dimensions of social CRM and RQ. Full article

Structured science summaries or research contributions using properties or dimensions beyond traditional keywords enhance science findability. Current methods, such as those used by the Open Research Knowledge Graph (ORKG), involve manually curating properties to describe research papers’ contributions in a structured manner, but this is labor-intensive and inconsistent among human domain-expert curators. We propose using Large Language Models (LLMs) to automatically suggest these properties. However, it is essential to assess the readiness of LLMs like GPT-3.5, Llama 2, and Mistral for this task before their application. Our study performs a comprehensive comparative analysis between the ORKG’s manually curated properties and those generated by the aforementioned state-of-the-art LLMs. We evaluate LLM performance from four unique perspectives: semantic alignment with and deviation from ORKG properties, fine-grained property mapping accuracy, SciNCL embedding-based cosine similarity, and expert surveys comparing manual annotations with LLM outputs. These evaluations occur within a multidisciplinary science setting. Overall, LLMs show potential as recommendation systems for structuring science, but further fine-tuning is recommended to improve their alignment with scientific tasks and mimicry of human expertise. Full article

Active learning approaches, incorporating student engagement through experimentation and problem solving, effectively foster higher-level thinking abilities and enhance academic performance. Interactive tools like simulators align with these methodologies, but commercially available simulators have limitations; particularly, their high cost and lack of customization features pose significant challenges for many educational institutions. This article presents CORES, a web-based educational application designed to simulate controlled rectifier circuits. CORES eliminates the need for intricate circuit assembly and software installation by providing pre-built circuits so that users can concentrate on analyzing circuit behavior by manipulating the thyristor firing angle and load characteristics, while the application generates output voltage and current waveforms under steady-state conditions, minimizing computation time. CORES has proven to be a valuable pedagogical tool, surpassing commercial simulators in terms of accessibility, ease of use, and enriched learning experiences for power electronics students and educators. Full article

Machine learning (ML) has become integral in educational decision-making through technologies such as learning analytics and educational data mining. However, the adoption of machine learning-driven tools without scrutiny risks perpetuating biases. Despite ongoing efforts to tackle fairness issues, their application to educational datasets remains limited. To address the mentioned gap in the literature, this research evaluates the effectiveness of four bias mitigation techniques in an educational dataset aiming at predicting students’ dropout rate. The overarching research question is: “How effective are the techniques of reweighting, resampling, and Reject Option-based Classification (ROC) pivoting in mitigating the predictive bias associated with high school dropout rates in the HSLS:09 dataset?" The effectiveness of these techniques was assessed based on performance metrics including false positive rate (FPR), accuracy, and F1 score. The study focused on the biological sex of students as the protected attribute. The reweighting technique was found to be ineffective, showing results identical to the baseline condition. Both uniform and preferential resampling techniques significantly reduced predictive bias, especially in the FPR metric but at the cost of reduced accuracy and F1 scores. The ROC pivot technique marginally reduced predictive bias while maintaining the original performance of the classifier, emerging as the optimal method for the HSLS:09 dataset. This research extends the understanding of bias mitigation in educational contexts, demonstrating practical applications of various techniques and providing insights for educators and policymakers. By focusing on an educational dataset, it contributes novel insights beyond the commonly studied datasets, highlighting the importance of context-specific approaches in bias mitigation. Full article

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