UROP Proceedings 2022-23

School of Engineering Department of Computer Science and Engineering 114 Assess User Experience to Design Effective Visual Representation and Interaction in Virtual Reality Supervisor: MA, Xiaojuan / CSE Student: JIA, Feiyu / DSCT Course: UROP2100, Fall Interaction is one of the key aspects of user experience in VR. To improve retention and enhance the user experience, the design of interaction should be natural and intuitive. Besides the traditional physical controller, hand gesture interaction is now applied in more VR projects aiming to provide a natural user experience with visual representation and interaction in immersive environments. With the help of developing toolkits, the design of various hand interactions beyond imagination is now accessible in VR apps and games in a new generation. This project focuses on the usability of hand gestures in immersive data visualization and explores the edge of accuracy, operability, and scalability of current recognition technology. Interaction Design for Human-AI Collaboration Supervisor: MA, Xiaojuan / CSE Student: BAI, Yihan / COMP XU, Minrui / DSCT Course: UROP1100, Fall UROP1100, Fall With the advances in AI technology, AI is increasingly adopted in numerous domains to assist humans in handling various tasks. However, AI may only achieve a perfect solution to some problems. Thus, human-AI collaboration is considered a suitable paradigm to leverage the respective capabilities of humans and AI to accomplish challenging tasks jointly. However, many important issues need to be solved in the collaborative process, including but not limited to whether and when people should trust AI, how humans and AI should appropriately divide tasks, etc. To tackle these problems, this project aims to design human-AI interactive interfaces to promote more transparent communication and effective collaboration between people and AI so that the human-AI team will achieve optimal performance. We use 2018 ACS income data tables from different states and GTB (Gradient Tree Boosting) to train 3 base models and apply stacking to ensemble the base models to achieve a complementary performance. After that, we will design interfaces and conduct user study. Interaction Design for Human-AI Collaboration Supervisor: MA, Xiaojuan / CSE Student: GUO, Meichen / COGBM Course: UROP1100, Summer With the appearance of ChatGPT, AI has again become a hot topic in the high-tech world, leading researchers on human-computer interaction to reconsider the full application of the potential from both sides to obtain optimal results in decision-making. Through the Undergraduate Research Opportunities Program, under the guidance of Professor Xiaojuan MA and Ph.D. student Shuai MA, I went deeply into the Interaction Design for Human-AI Collaboration. Specifically, I read many authoritative papers from academic conferences like CHI, IUI, and CSCW to grasp the latest research achievements, and participated in the design of Human SelfConfidence Calibration, proposing ideas and analyzing data. Looking ahead, I will continue my involvement in the project and make further contributions to data analysis and paper writing.