Authors - Madhusmita Chakraborty, Vijay Kumar Nath, Deepika Hazarika Abstract - Due to morphological similarities between species, environmental variability, and the requirement for specialized knowledge, accurate identification of medicinal plants is still difficult, despite their critical role in primary healthcare systems around the world. A Hybrid RegNetX-Hierarchical Bidirectional Linear Cross Attention framework referred to as HR-HBCA framework for identifying medicinal plants from leaf photos is presented in this work. Multi-scale features are extracted using a RegNetX backbone, and computationally efficient linear crossattention is used in Hierarchical Bidirectional Linear Cross-Attentive Fusion (HBLCAF) blocks to integrate shallow spatial and deep semantic representations. Balanced contextual exchange across scales is achieved by bidirectional cross-attentive fusion. The HR-HBCA framework shows strong performance under notable intra-class variability, with accuracies ranging from 93.79% to 99.73% when tested on five diverse public datasets.
Authors - Ntima Mabanza Abstract - Research that examines the use of Pedagogical interface agents (PIAs) in digital learning environments has demonstrated that PIAs can increase learner engagement, motivation, knowledge retention, and improve the learning outcomes. Despite that, there is limited empirical understanding of which PIA’s particular features are very noticeable and preferred by learners. A mixed-methods approach was used in this study, combining initial training, task completion, and feature rating questionnaires with 62 adult participants. This approach was used to examine adult learner preferences for PIAs’ noticeable features, such as appearance, voice, and movement. The study findings indicate that adult learners prioritize PIAs’ movement, followed by their appearance, and lastly their voice. The findings of this study provide very useful design guidelines for developing effective learner-centered PIA systems that maximize engagement and satisfaction.
Authors - Imene Kichah, Amir Aieb, Antonio Liotta, Muhammad Azfar Yaqub Abstract - The rapid growth of Information and Communication Technologies (ICT) has profoundly altered educational systems by redefining teaching practices, institutional processes, and professional expectations. Within the broader context of sustainable development and smart education, ICT has emerged as an important facilitator of efficiency, accessibility, and innovation. This paper presents a conceptual analysis of how ICT can contribute to sustainable development through its influence on teachers’ work–life balance and job satisfaction in ICT-enabled learning environments. While ICT adoption has the potential to enhance instructional flexibility, autonomy, and efficiency, excessive digital connectivity, intensified workload, and blurred work–life boundaries may adversely affect teachers’ well-being. The paper identifies work life balance as a key mediating factor linking ICT use to job satisfaction and long term professional sustainability. Furthermore, the study situates teachers’ well being within the broader framework of sustainable development, emphasizing its relevance to Sustainable Development Goals such as SDG 3 (Good Health and Well-Being), SDG 4 (Quality Education), and SDG 8 (Decent Work and Economic Growth). The analysis underscores the need for human-centred, policy-driven, and ethically oriented ICT integration strategies that prioritize teacher well-being alongside technological advancement. The paper contributes to the discourse on sustainable and intelligent education systems by highlighting that the long-term effectiveness of ICT-driven educational transformation depends on balanced digital practices that support teachers’ work–life balance and job satisfaction.
Authors - Aleah Prameswari Kalyana Merkadea Purnomo, Muhammad Aras Abstract - TikTok Live Shopping has been rapidly growing and the way consumers and brands interact has changed, with emotional and communicative engagement leading the way to driving purchases. However, there is minimal literature to understand the impact of how host performance, emotional euphoria, and perceived quality value combine to affect impulse buying, specifically in reference to preloved fashion and the Generation Z cohort. This study aims to fill the gap in literature by examining the impact of these three components on impulse buying behavior from the perspective of Integrated Marketing Communication (IMC). In this study, a quantitative method was used by conducting an online survey with 136 respondents from Generation Z who have bought items through TikTok Live Shopping. The data was analyzed using Partial Least Squares–Structural Equation Modeling (SEM-PLS). Emotional euphoria is the only antecedent with a statistically significant positive relationship with impulsive buying behavior. Host performance and quality value have a positive relationship but are statistically insignificant. Moderately, the model explains 57% of the variance in impulsive buying (R² = 0.570) showing moderate predictive power. Emotional stimulation is the largest driver of im-pulsive buying, while cognitive evaluation centered around quality is merely justifying a post purchase rationale. This paper illustrates that in live commerce, emotional irrationality is more dominant than communicative rationality, offering a new dimension to the IMC paradigm in the context of real-time social commerce and underlining the criticality of emotional engagement in live sessions for improving customer conversion.
Authors - Matthew Abrham Kristanto, La Mani, Cindy Magdalena, Maudi Aulia Saraswati, Annisa Atha Hanifah Abstract - Digital Twins (DTs) are increasingly explored for integrating BIM and IoT in facility management, yet many implementations remain fragmented, weakly governed semantically, and difficult to scale. This paper presents a BIM-centric DT framework for the MaCA museum Living Lab in Turin, combining indoor–outdoor environmental sensing, automated BIM synchronization, IFC-based interoperability, and a prototype temporal analytics layer. The methodology links shared-parameter modeling, Dynamo–Python synchronization, and room-/zone-level identifier logic to validate end-to-end snapshot-to-BIM integration on a one-week monitoring dataset. Results confirm robust parameter mapping, successful serialization of custom space-level IFC property sets, and the feasibility of a dual-layer DT strategy in which BIM/IFC supports semantic-spatial contextualization while external temporal platforms support analytics and dashboard visualization. The core contribution lies in defining a scalable and standards-aligned workflow for cultural facilities based on identifier persistence, modular synchronization, interoperability, and data-quality-aware DT deployment.
Authors - Md. Abdul Malek Sobuj, Md. Faruk Abdullah Al Sohan, Afroza Nahar, Saeeda Sharmeen Rahman Abstract - Tomato leaf diseases lead to significant losses in yield and quality, especially in developing areas where timely diagnosis and expert help are scarce. Early and accurate disease detection is vital for sus tainable crop protection and better agricultural productivity. This pa per proposed a hybrid AI-IoT imaging framework for early-stage multi label tomato leaf disease detection in real-field agricultural settings. The proposed hybrid framework combines camera-based IoT sensing, edge and cloud computing, and a lightweight hybrid CNN, the Transformer model, to allow continuous monitoring, timely diagnosis, and decision support. The proposed hybrid framework merges local feature extrac tion with global context modeling to enable accurate multi-label clas sification while being suitable for deployment on devices with limited resources. A conceptual performance comparison and case study show the practical feasibility and benefits of this approach regarding diagnos tic reliability, scalability, and cost-effective deployment. The framework aims to improve early disease identification, reduce crop losses, and sup port precision agriculture practices. This study offers a practical and scalable solution for intelligent tomato disease management and aids the development of sustainable AI-IoT-based smart agriculture systems.
Authors - Fahima Sultana Smrity, Md. Ibrahim Tanjim, Md. Faruk Abdullah Al Sohan, Afroza Nahar, Saeeda Sharmeen Rahman Abstract - Solar-powered systems in railway crossing safety are an effi cient approach for ensuring continuous monitoring and accident preven tion in risky and less supervised areas. Solar energy ensures the reliability of the system, while the components connected to it are optimized for en ergy efficiency and long-range communication. In the transportation sec tor, IoT-enabled safety devices are gaining importance, and railway cross ings are a key example. This paper proposes a simplified solar-powered model, called Smart Railway Crossing Protection (SRCP), for railway au tomation using IoT-based sensing and communication. This model intro duces an energy-efficient design with LiFePO4 battery backup, MPPT based solar adaptation, and wireless communication of the LoRa model, focusing on reducing functional costs and dependence on manual su pervision compared to traditional railway safety systems. The proposed system aims to increase real-time responsiveness, ensure stability in re mote places, and improve the overall security of the passenger and vehi cle. Moreover, the SRCP model emphasizes scalability and adaptability, underlining its importance for various railway infrastructures.
Authors - Subhrajyoti Sunani, Prasant Kumar Sahu, Debalina Ghosh Abstract - Topic detection is an essential task in Natural Language Processing (NLP) that enables the automatic classification of text into predefined categories. However, research challenges in the Myanmar language remain limited due to the lack of annotated corpora and linguistic challenges. In this study, word-level segmentation is employed to capture more semantically meaningful units for topic detection, such as အနုပညာ (art), ဥပဒေ (law), အာားကစာား (sports), and နည ားပညာ (technology). The study trains and evaluates the system on a dataset of News articles categorized into 12 predefined topics: agriculture, art, crime, disaster, economy, education, foreign affairs, health, politics, religion, sports, and technology. A variety of models was examined, covering traditional machine-learning baselines, a deep learning sequence model, and transformer-based architectures. Logistic Regression and Naïve Bayes were tested and achieving accuracies of 0.73 and 0.63, respectively, with Logistic Regression outperforming Naïve Bayes as a stronger linear baseline. The LSTM model, which incorporates sequential dependencies, improves performance further with an accuracy of 0.85. Transformer based approaches deliver the best results: DistilBERT achieves 0.87 accuracy, while word level mBERT achieves 0.95 accuracy at its peak, demonstrating the effectiveness of word-level approaches for Myanmar topic detection. Overall, the findings demonstrate that while traditional models offer useful baselines, deep learning and especially transformer-based architectures provide substantial gains in accuracy and reliability for Myanmar topic detection. This research highlights the effectiveness of modern transformer-based methods for low resource language applications and sets a benchmark for future work in Myanmar NLP.
Authors - Ayana Soman, Diya P. Varghese, Elizabeth Anna Liju, Ethel Jimmy, Liyan Grace Shaji, P R Neethu Abstract - Radiology report generation is a vital and time-consuming part of medical imaging workflows. It is often shaped by heavy workloads and differences in opinions among observers. This paper presents RadVi sion, an AI-driven platform designed to automatically generate prelimi nary radiology reports from medical imaging data, with a specific focus on MRI scans. The framework uses Vision Transformers (ViT) for global feature extraction and Topological Data Analysis (TDA) to identify structural and shape-based abnormalities that traditional deep learning methods might miss. To improve understanding and clinical reliability, RadVision includes explainability tools like Grad-CAM heatmaps and persistence diagrams from TDA. A transformer-based language model creates structured, editable diagnostic reports with confidence scores, allowing for effective validation by humans. The system is accessible through a secure web dashboard, facilitating collaborative annotation, feedback-based model improvement, and smoother workflow integration. Experimental tests across various radiological cases show better diagnos tic support, greater transparency, and less reporting effort. These results position RadVision as a scalable and clear AI tool to assist radiologists and promote efficient and reliable medical reporting.
Authors - Aura Meivia Safira Arsya, Ricardo Indra, Shafa Salsabila Risfi Febrian, Benedicta Kalyca Kyatimanyari Abstract - This study analyzes the extent to which credibility from influencers impacts consumers' buying behavior. The focus will be on how the intention to buy impacts this relationship as the problem is being analyzed in the context of social commerce on TikTok. The study is developed within the framework of Source Credibility Theory which suggests that consumers’ perception and consequent behavior are influenced by the perceived degree of the spokesperson’s Attractiveness, Trustworthiness, and Expertise. The study employs a quantitative explanatory methodology. A purposive sampling technique was used to collect data from a sample of 100 active TikTok users who follow the provided influencer. The analyzed relationships will be quantified using Structural Equation Modelling with Partial Least Squares (SEM-PLS). The research results concluded that influencer credibility increases the intention to buy, but does not increase the purchasing decision. The intention to buy completely mediates the relationship between influencer credibility and purchasing decision. This demonstrates that influencer credibility is a significant factor in the intention to buy behavior, but it is the intention that is essential in order to convert the persuasive influence into actual buying behavior. The study contributes to digital marketing communication research by extending Source Credibility Theory to the context of short-video social commerce platforms.
