EAI Endorsed Transactions on Context-aware Systems and Applications

Elevating User-Centered Design with AI: A Comprehensive Exploration using the AI-UCD Algorithm Framework

Waralak Vongdoiwang Siricharoien — 2024-03-15

This paper presents a comprehensive exploration of the synergistic relationship between User-Centered Design (UCD) and Artificial Intelligence (AI) within the context of the AI-UCD Algorithm Framework. With the growing influence of AI in digital interfaces, the need to prioritize user needs and preferences has become paramount. The AI-UCD Framework, consisting of nine pivotal steps, acts as a structured guide for integrating AI into user interfaces while ensuring a user-centric, data-driven, and ethical approach. The exploration begins by highlighting the importance of understanding user needs and context through robust user research and contextual inquiry. It then delves into the process of defining AI integration objectives and brainstorming AI-enhanced solutions, emphasizing the creative aspects of UCD in tandem with AI capabilities. Subsequently, the paper discusses the critical role of designing AI-driven interfaces, from information architecture to user flow design, ensuring seamless integration of AI features.Implementation and testing of AI features are addressed, highlighting the collaboration between UI/UX designers and AI developers. The paper emphasizes the iterative nature of the framework, relying on usability testing and user feedback to drive continuous improvements. Moreover, it considers user training and assistance, a vital aspect of introducing users to AI features.The framework's data-driven aspect is covered by discussing data collection, analysis, and performance monitoring to ensure AI features are meeting objectives and KPIs. Additionally, the exploration addresses AI's role in personalization, adapting to user behavior and preferences. It recognizes the ethical dimensions of AI, promoting transparency, fairness, and accessibility.The paper then presents a five-step AI-UCD Validation Model, designed to verify the framework's effectiveness in real-world applications. These validation steps encompass user testing and feedback, data analysis, ethical audits, iterative improvements, and compliance with industry standards. Examples of how these steps work in practice are provided.

Integration and Recommendation System of Profiles based on Professional Social Networks

Paul Dayang — 2024-01-15

The aim of our investigation is to personalize bilateral recommendation of job-related proposals based on existing professional social networks. In a context where the points of view of job seekers and employers can be contradictory, our approach consists in trying to bring the both in a best possible matching. To this end, we propose an integration system that gives a minimum of credit to the users’ data in order to facilitate the discovery of relevant proposals based on the users’ behaviors, on the characteristics of the proposals and on possible relationships. The main contribution is the proposal of an architecture for the recommendation of profiles and job offers including social and administrative factors. The particularity of our approach lies in the freedom from the recommendation problem by using metrics proven in the literature for the estimation of similarity rates. We have used these metrics as default values to appropriate data dimensions. It emerges that, the user’s behavior is exclusively responsible for the recommendations. However, the cross-analysis of randomly generated behaviors on real profiles collected on Cameroonian sites dedicated to job offers, shows the influence of the most active users. But, for requests via the search bar (interface with the script respecting the path of our architecture) the central subject remains the user. Our current work is limited by a data set that is not very representative of changing socio-economic conditions.

UGGNet: Bridging U-Net and VGG for Advanced Breast Cancer Diagnosis

Tran Cao Minh — 2024-01-12

In the field of medical imaging, breast ultrasound has emerged as a crucial diagnostic tool for early detection of breast cancer. However, the accuracy of diagnosing the location of the affected area and the extent of the disease depends on the experience of the physician. In this paper, we propose a novel model called UGGNet, combining the power of the U-Net and VGG architectures to enhance the performance of breast ultrasound image analysis. The U-Net component of the model helps accurately segment the lesions, while the VGG component utilizes deep convolutional layers to extract features. The fusion of these two architectures in UGGNet aims to optimize both segmentation and feature representation, providing a comprehensive solution for accurate diagnosis in breast ultrasound images. Experimental results have demonstrated that the UGGNet model achieves a notable accuracy of 78.2\% on the "Breast Ultrasound Images Dataset."

A hybrid classification model in improving the classification quality of network intrusion detection systems

Thanh Hoàng Ngọc — 2025-06-17

Stream-based anomaly detection is an issue that continues to be researched in the cybersecurity environment. Much previous research has applied machine learning as a method to improve anomaly detection in network intrusion detection systems. Recent research shows that network intrusion detection systems still face challenges in improving accuracy, reducing false alarm rates, and detecting new attacks.

The article proposes a hybrid classification model that combines improved data preprocessing techniques with ensemble techniques. Experimental results on the UNSW-NB15 dataset show that the proposed solutions have helped improve the classification quality of network intrusion detection systems compared to some other research.

Analyzing online reviews at the word level to understand customer experience

Ha Thi Thu Nguyen — 2025-07-08

INTRODUCTION: In the competitive business environment, customer experience plays a pivotal role in driving brand success. Brands that deliver exceptional customer experiences benefit from increased loyalty, advocacy, and stronger market differentiation. With the rise of digital platforms, customers frequently share post-purchase experiences online, making sentiment analysis essential for strategic marketing.

OBJECTIVES: This study aims to explore customer experience with the Trung Nguyen Legend coffee brand by analyzing user-generated content on TripAdvisor. It seeks to identify key aspects of customer feedback and measure satisfaction and loyalty levels.

METHODS: The research employs natural language processing (NLP) techniques and Python-based sentiment analysis tools. Specifically, aspect-based sentiment analysis (ABSA) is used to extract and evaluate sentiment associated with different service dimensions based on online reviews.

RESULTS: The analysis reveals that Trung Nguyen Legend achieves a Customer Satisfaction (CSAT) score exceeding 66% and a Net Promoter Score (NPS) over 34%. These results indicate a generally positive customer experience, with specific strengths and areas for improvement clearly identified.

CONCLUSION: The study demonstrates that ABSA is a cost-effective and time-efficient method for understanding customer sentiment. The findings offer valuable insights for enhancing customer experience management and inform strategic improvements for the Trung Nguyen Legend brand.

Human Emotion Recognition with an Advanced Vision Transformer Model

Kha Tu Huynh — 2025-04-30

This paper proposes a novel deep-learning technique that leverages the Efficient Vision Transformer –M5 (Efficient ViT-M5) model to improve the existing design by offering a more computationally economical version that maintains good performance, making it highly suitable for practical applica-tions. The utilization of transfer learning involved leveraging pre-trained weights from the ImageNet dataset, substantially enhancing the model's accu-racy and efficiency. The proposed method involves training the advanced Effi-cientViTM5 model utilizing three widely recognized facial emotion recognition datasets: FER2013+, AffectNet, and RAF-DB. A comprehensive data augmentation pipeline is employed to enhance the diversity of the training data and bolster the model's robustness. The trained proposed model proved exceptional accuracy rates of 94.28% (FER2013+), 94.69% (AffectNet), and 97.76% (RAF-DB). The results emphasize the strength and effectiveness of the proposed model in identifying face emotions in various datasets, showcasing its potential for practical use in emotion-aware computing, security, and health diagnostics. The research significantly improves facial emotion recognition by introducing a reliable and practical way of recognizing emotions using cutting-edge deep learning techniques. The results show the possibility of enhancing and flexible interactions between humans and computers, highlighting the efficacy of sophisticated deep learning models in addressing complex computer vision problems.

Temporal assessment of cognitive load factors using ocular features during a visual search

Minoru Nakayama — 2025-08-05

The possibility of evaluating temporal changes in cognitive workloads during a visual search task is examined using microsaccade (MS) rates and pupillary changes. The experimental task was designed as a search for a speciﬁc ﬁgure, where task diﬃculty and reaction accuracy during the trials were controlled. Individual cognitive workloads were measured after the experimental sessions were conducted, using NASA-TLX scale ratings. Temporal changes in the cognitive load were identiﬁed using metrics of oculomotors during two stages of task processing, by comparing cognitive loads with individual ratings on a scale. Since the source of the load may be a common one, changes in latent attention resources required for the task were estimated with a designated state-space model, using the observation data in order to synthesise measurement of MS rates and pupillary changes. The predicted levels of attention resources correspond to the activity during the performance of the experimental tasks during the trials, and reﬂected some of the rating scores for workload scales. Also, the ranges of conﬁdence intervals for attention resources correlate signiﬁcantly with the ratings for information processing at the stage where visual stimulus is presented during tasks.

Reflecting Society Through Art: The Aesthetic Evolution of Paintings

Nattanun Siricharoen — 2025-07-22

The COVID-19 pandemic has sparked a powerful artistic response in the form of a captivating painting series, offering a compelling reflection of society during this unparalleled crisis. Comprising five distinctive pieces, these artworks provide a poignant glimpse into humanity's collective response to the pandemic, encapsulating emotions ranging from fear and anxiety to resilience and adaptability. Central to this artistic narrative is the face mask, an innovation born out of necessity that has come to symbolize humanity's resolve to adapt for survival. The mask represents not only our need for protection but also serves as a tangible reminder of our shared struggle and our capacity to respond to adversity with creativity and unity. Through this artistic endeavor and subsequent exhibition, the artist skillfully navigates the complexities of human existence. Life, as depicted in these artworks, is a tapestry woven with threads of happiness and suffering, where smiles and tears coexist harmoniously, reflecting the intricate truth of our journey. These paintings compel us to confront life's challenges and adapt, mirroring the collective response to the pandemic. They remind us that, as individuals and as a society, we possess the resilience to persevere and evolve in the face of adversity. This series is a testament to the indomitable human spirit, showcasing our ability to find strength and beauty even amidst the most trying circumstances. In summary, this artistic painting series is a profound exploration of the human experience during the COVID-19 pandemic. It encapsulates our shared emotions and the significance of adaptation as a vital element of survival. These artworks serve as a poignant reminder that, no matter the challenges we encounter, we have the inner resources to endure, thrive, and discover moments of profound beauty amidst adversity.

Augmented Reality and Virtual Reality Crafting New Dimensions of Human Perception, Interaction, and Societal Evolution

Waralak Vongdoiwang Siricharoen — 2025-05-27

The rapid evolution of Augmented Reality (AR) and Virtual Reality (VR) is not just an expansion of human-computer interaction but a radical redefinition of human existence itself. This paper delves into the extraordinary potential of AR and VR to reshape reality, alter consciousness, and create immersive, parallel dimensions. As AR/VR technologies blur the boundaries between the physical and digital realms, they are poised to redefine industries, social structures, and even the nature of human cognition. We explore how AR/VR could soon transcend entertainment and education, emerging as the cornerstone of a hybrid reality where the digital and physical worlds are indistinguishable. Furthermore, we speculate on their role in crafting synthetic worlds, which challenge our notions of reality, identity, and interaction, and question what it means to exist in an era where the virtual is as real as the tangible.

Efficient Key Frame Extraction from Videos Using Convolutional Neural Networks and Clustering Techniques

Anjali H Kugate — 2024-07-17

One of the most reliable information sources is video, and in recent years, online and offline video consumption has increased to an unprecedented degree. One of the main difficulties in extracting information from videos is that unlike images, where information can be gleaned from a single frame, a viewer must watch the entire video in order to comprehend the context. In this work, we try to use various algorithmic techniques, such as deep neural networks and local features, in conjunction with a variety of clustering techniques, to find an efficient method of extracting interesting key frames from videos to summarize them. Video summarization plays a major role in video indexing, browsing, compression, analysis, and many other domains. One of the fundamental elements of video structure analysis is key frame extraction, which pulls significant frames out of the movie. An important frame from a video that may be used to summarize videos is called a key frame. We provide a technique that leverages convolutional neural networks in our suggested model, static video summarization, and key frame extraction from movies.

Toward Modeling Linguistic Fuzzy Spanning Trees Based on Hedge Algebra

Nguyen Van Han — 2024-09-27

This paper presents an innovative approach to modeling Linguistic Fuzzy Maximum Spanning Trees (L-FMSTs) using Hedge Algebra (HA). HA provides a robust framework for quantifying linguistic terms, which is essential for handling the vagueness inherent in natural language. By integrating HA with L-FMSTs, we aim to enhance the interpretability and performance of fuzzy systems in applications requiring complex decision-making and optimization.

Fuzzy Message Passing in Graph Neural Networks: A First Approach to Uncertainty in Node Embeddings

Minh Tuan Duong — 2025-07-15

Graph Neural Networks (GNNs) have emerged as a powerful tool for learning representations in graph structured data. However, traditional message-passing mechanisms often struggle with uncertainty and noise in node features and graph topology. In this paper, we propose Fuzzy Message Passing (FMP), a novel approach that integrates fuzzy max-min aggregation into GNNs to improve robustness against uncertainty. Our method enhances node embeddings by leveraging fuzzy logic principles, ensuring better stability and interpretability in complex graph tasks. Experimental results on benchmark datasets demonstrate that FMP outperforms conventional message-passing schemes, particularly in scenarios with noisy or incomplete data.

Fuzzy Graph Neural Networks: A Comprehensive Review of Uncertainty-Aware Graph Learning

Ngoc Dan Tran — 2025-07-16

Graph Neural Networks (GNNs) have become powerful tools for learning from graph-structured data. However, traditional GNNs often fail to address uncertainty inherent in many real-world applications. Fuzzy Graph Neural Networks (FGNNs) integrate fuzzy logic into GNNs to provide a robust mechanism for managing uncertainty, imprecision, and vagueness. This paper presents a comprehensive review of FGNNs, examining their theoretical underpinnings, methodologies, applications, challenges, and potential research directions.

A Review of Quantum Lambda Calculi: Linearity, Semantics, and Programming Models

Tran Ngoc Dan — 2025-07-17

Quantum lambda calculi extend classical lambda calculus to model quantum computation by integrating linear types, quantum operations, and classical control. This paper surveys key calculi—including QΛ, QLC, Proto-Quipper, and QML—highlighting their design principles, type systems, and semantic foundations. By comparing their approaches to handling quantum data, control flow, and circuit construction, we provide insights into the current state and future directions of quantum programming language research.

A Survey of Quantum Type Theory: From Linearity to Formal Verification

Van Han Nguyen — 2025-07-16

Quantum Type Theory (QTT) provides a formal system that combines ideas from quantum mechanics, type theory, and logic to support reliable quantum programming. Since quantum information cannot be copied or deleted like classical data, QTT uses linear types to ensure that quantum operations follow the laws of physics. This paper reviews the main concepts in QTT, such as linear functions, tensor products, and dependent types, and explains how they help programmers write safe and correct quantum code.

Tracing the Evolution of Max-Min Aggregation and Fuzzy Systems in AI: A Bibliometric Review

Nguyen Van Han — 2025-07-22

This paper presents a bibliometric review of Max-Min aggregation functions and fuzzy systems in artificial intelligence (AI) from 1990 to 2024. Drawing on data from Scopus and analyzed using Bibliometrix and VOSviewer, we map publication trends, key contributors, thematic developments, and emerging research areas. The findings reveal growing interest in interpretable AI, neuro-fuzzy models, and hybrid systems. We highlight the integration of Max-Min aggregation in explainable AI and identify key research gaps. This review provides a structured overview of the field’s evolution and offers guidance for future research directions.

Max-Min Aggregation in Fuzzy Linguistic Systems and Machine Learning: A Narrative Review

Nguyen Van Han — 2025-07-22

Max-min aggregation functions play a fundamental role in fuzzy linguistic systems and machine learning by providing interpretable and mathematically sound methods for combining imprecise and qualitative information. This narrative review synthesizes the key concepts, models, and applications of max-min aggregation, highlighting its significance in enabling human-centric reasoning and explainable AI. We discuss theoretical foundations, linguistic modeling frameworks, and diverse practical applications, including decision support systems and fuzzy rule-based classifiers. Challenges such as scalability, integration with deep learning, and semantic standardization are identified, along with promising future research directions. This review aims to provide a comprehensive understanding of max-min aggregation’s contributions to interpretable and flexible AI systems

Systematic Review of Max-Min Aggregation in Fuzzy Systems and Interpretable Machine Learning: Models, Evaluation, and Applications

Nguyen Van Han — 2025-07-22

This systematic review investigates the use of max-min aggregation in fuzzy systems and interpretable machine learning. Rooted in fuzzy set theory and triangular norms, max-min aggregation offers a transparent and mathematically simple approach to modeling uncertainty and decision-making. We examine theoretical foundations, practical applications, evaluation methods, and comparative taxonomies. The review identifies key challenges such as scalability and integration with learning algorithms, and highlights future directions for improving transparency in AI. Our findings underscore the relevance of max-min aggregation in developing interpretable and responsible AI systems.