Welcome to my research portfolio. This space is dedicated to exploring the intersection of design, technology, and human behavior. Here, you'll find my work spanning various projects, from user experience studies to machine learning applications, aimed at improving user interactions and making technology more intuitive. Through these research efforts, I seek to contribute to the broader conversation about how design can shape the future of human-computer interaction.
Advancements in Social Network Analysis through Graph Neural Networks
Abstract
The recent integration of Graph Neural Networks (GNNs) into social network analysis represents a significant advancement in the field of data science. Social networks, defined by intricate relationships and user interactions, pose unique challenges that conventional analytical methods often fail to overcome. GNNs utilize the inherent graph structure of these networks, allowing for the effective modeling of nodes (representing individuals) and edges (representing relationships) to uncover complex patterns and dynamics. This paper examines the application of GNNs across various facets of social network analysis, such as community detection, influence propagation, and user behavior prediction. By employing GNNs, researchers can improve prediction accuracy and gain deeper insights into social structures, facilitating more effective engagement and intervention strategies. Additionally, we address the scalability of GNNs in managing large datasets typical of social networks and their capability for real-time analysis. This study highlights the importance of GNNs as a robust tool in the rapidly evolving landscape of data science, providing innovative solutions to the challenges inherent in social network analysis.
Abstract
The rise of online grocery shopping has created a strong need for user-friendly digital solutions. In this study, we look closer at E-Hut, an e-commerce app that makes grocery shopping more accessible and enjoyable with its simple design and features designed with users in mind. Focusing on the user's needs, we explore how effective E-Hut's key features—like the shopping list, special deals, and notification systems—are in improving the grocery shopping experience. We also assess how E-Hut affects user satisfaction and retention through in-depth UX research, surveys, and hands-on testing. Ultimately, we position E-Hut as a shining example for future e-commerce innovations.
Abstract
This study delves into PantBee, a newly developed mobile application aimed at helping gardening enthusiasts with plant care and garden management. By leveraging augmented reality (AR) and artificial intelligence (AI) for personalized plant care recommendations, along with e-commerce functions and a reminder system, PantBee enhances the gardening experience. Spanning 15 weeks, the development process included user interviews, persona creation, and usability testing, leading to over 60 screens and 10 standout features. Insights from 22 user interviews shaped key design elements, dramatically improving user satisfaction and engagement while simplifying essential gardening tasks. Notably, users reported a 30% improvement in managing garden spaces, a 25% reduction in plant neglect, and a 20% decline in plant mortality rates, thanks to AI-driven care recommendations.
Abstract
The complexity of international university applications presents numerous challenges for students, including unclear requirements, intricate visa procedures, and scholarship eligibility confusion. This study examines Uniway, a mobile application developed as a digital agency platform that supports students in their higher education journeys. Integrating extensive user experience (UX) research and leveraging the Design Thinking framework, Uniway provides features for admission resources, visa requirements, scholarship filtering, and expert consultation. Through over 150 user surveys, the design and development team created 85 screens and 60 reusable components, ensuring adaptability to evolving needs. Evaluation results show a 35% reduction in stress levels, a 45% improvement in locating critical information, and a 30% decrease in navigation errors during scholarship and document submission tasks. This paper documents the iterative development of Uniway and evaluates its impact, providing a model for similar educational technologies.