Exploring the diversity of retweeting behavior patterns in Chinese microblogging platform. Issue 4 (July 2017)

• Record Type: Journal Article
• Authors: Li, Qianqian; Liu, Yijun
• Is Part Of: Information processing & management. Volume 53:Issue 4(2017:Jul.)
• Page Start: 945
• Page End: 962
• Publication Date: 2017-07
• Subjects: Retweeting behavior -- Time series -- Pattern discovery -- Classification; Information storage and retrieval systems -- Periodicals; Information science -- Periodicals; Systèmes d'information -- Périodiques; Sciences de l'information -- Périodiques; Information science; Information storage and retrieval systems; Periodicals; 658.4038
• DOI: 10.1016/j.ipm.2016.11.001
• Languages: English
• ISSNs: 0306-4573

Abstract

Highlights: Proposing a new time series reduction method to capture the variations of retweeting time series at the structural level. Identifying representative retweeting temporal patterns. Conducting comprehensive analysis for different classes. Presenting a classification model predicting retweeting temporal patterns. Abstract: Information regarding retweeting behavior patterns is crucial for understanding online information diffusion, product promotion, and other social contagion dynamics. In this study, we explored the retweeting behavior patterns of influential microblogs according to detailed retweeting behavior data gathered from Sina Weibo, a Twitter-like online social network in China. We devised an innovative time series reduction method to capture retweeting time series features at the structural level, and observed three types of daily representative temporal patterns and six types of hourly representative temporal patterns. We then conducted comprehensive analyses of each pattern, including the statistical features of the time series and other factors influencing the formation of patterns (e.g., posting users, content, and publication time) of microblogs. Based on readily available social-influential, topical, and temporal factors, we also established a classification model that relatively reliably predicts the temporal class of an original microblog's retweeting time series. The results presented here may offer insight into common temporal patterns of retweeting behavior on content-based online social networks.