Prediction of Posting Behavior of Online UGC Communities Users Based on Poisson Regression Xingyu CHEN a , Yu You LING a , Xianqi HU a , Xiuchen LIN a and Liang ZHU b1 a College of Management, Shenzhen University, China b College of Economics, Shenzhen University Abstract. As a new form of Internet service, User Generated Content (UGC) community has gained popularity these years, and also attracts academic attention. The posting behavior of users is an important index to measure whether the users are active or not. Identifying the potential active users in UGC Communities helps the companies make proper management strategies, which renders the prediction of users’ posting behavior important. This study used Poisson regression to examine the effect of users’ past behavior and status on users’ later behavior of posting. The data for analysis were extracted from ten thousand users on Qiushibaike throughout a window of three months, which included the number of posts and the number of likes, comments, and times they were featured on the home page by administrators. The results indicate that users’ past comments have a negative effect on users’ later posting behavior; users’ past posts have a positive effect on users’ later posting behavior; users’ past likes and hot posts they got have a positive effect on users’ later posting behavior. Keywords. UGC Online Community, Poisson Regression, Community Participation Introduction With the introduction of Web 2.0 15 years ago, online communities catering to User- Generated Content (UGC) is already a concept that marketers, consumers and scholars are familiar with. UGC is presented in many formats including video, still imagery, text, and even audio. These various content formats have become the focus of social media platforms big and small. In academia, UGC has also become a pertinent topic of research and has accumulated numerous discoveries. In this paper, User-Generated Content is classified as visual and audio web content that facilitates the sharing of personal opinions and knowledge. Through the acts of commenting, sharing, and liking such content, users interact with other members of the community [1][2][3]. Platforms like Facebook, Twitter, Instagram, Weibo, Tik-Tok, QiushiBaike have become household names, representing a group of UGC Communities that have garnered a significant and active user base. UGC Communities have the ability to concurrently provide the utility of learning, self-expression, and entertainment for large amounts of users. At the same time, while 1 Corresponding Author, Email: lzhu111@szu.edu.cn. Transdisciplinary Engineering for Complex Socio-technical Systems K. Hiekata et al. (Eds.) © 2019 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). doi:10.3233/ATDE190149 429