UROP Proceedings 2020-21

School of Engineering Department of Computer Science and Engineering 132 Analysis and Understanding of User Behaviors in Online Communities Supervisor: MA Xiaojuan / CSE Student: WU Zhiyang / COSC Course: UROP1100, Summer In 2020, the outbreak of COVID-19 in Wuhan, China caused a severe shortage of medical resources at the early stage. As a result, hospitals were struggling to cope with the volume of patients. Many COVID-19 patients (including suspected patients) who were not accepted or treated by hospitals, turned to social media sites for seeking assistance. In this project, we collect help-seeking posts from different online communities in mainland China, including Weibo, Toutiao, and some self-created websites, and analyze help-seeking behaviors during the outbreak of COVID-19 pandemic in Wuhan, China at the beginning of 2020, with a focus on the information collected from Weibo. Exploring College Students’ Self-identity via Social Media Supervisor: MA Xiaojuan / CSE Student: XU Mingshi / COMP Course: UROP1100, Spring COVID-19 has influenced our lives fundamentally. It also profoundly reconstructed the behavior of Chinese university students, as college students usually use social media, especially Weibo, to update their activities and express their feelings and opinions in real-time. Therefore, these contents reflect their daily life and their mental state. In addition, some researchers found that activities on social media networks (i.e. tagging and sharing) provide insights into the self-identity of users. Therefore, we start using social media content to study the impact of COVID-19 on college students 1 and compare the reaction between students in different identity groups. Our understanding of the impact of the pandemic on users’ social media engagement will facilitate identifying users vulnerable to crises and the subsequent provision of appropriate psychological support by social media. Gaining Insights from Noisy Data: Designing Transparent System for Online Decision Making Supervisor: MA Xiaojuan / CSE Student: CHEN Zixin / DSCT Course: UROP1100, Spring Users today have instant access online to large amounts of information on which to base their decisions. For example, they can choose products or book hotels based on user reviews or system recommendations. However, such information can be conflicting, biased, subjective and scattered across many sources. Solely relying on automatic approaches to address these issues can be insufficient due to inadequate usage of context and constantly updated users needs or criteria. This project aims to explore novel, transparent designs that increase user satisfaction, engagement and efficiency with actual decisions. We built interactive prototype systems and evaluate their effects in users decision-making process. Personally, I participated in the second half of the whole project and mainly responsible for the data collection, labeling and processes, web page & database construction, mock experiments and part of final results analysis.