[目的/意義]以微信用戶群分享數據為樣本,對用戶群體互動行為進行直接觀察,對于社交媒體中信息定向投放和輿情導控有著重要的意義。[方法/過程]通過構建APP平臺實現了微信用戶信息采集,并借助于人類動力學對微信用戶群內分享對象選擇、互動內容選擇和分享行為時間特征進行了分析。[結果/結論]研究表明微信中分享行為存在顯著的陣發性,群內用戶趨向于異配連接,但多數用戶是以旁觀者身份出現,高頻率互動僅停留在少數參與者層面,而帶有情緒誘導和相關利益引導標題的文章能夠喚起更多用戶的互動意愿。[局限]數據來源不夠廣泛,有可能導致研究偏差。

[Purpose/significance] Taking the sharing data of Wechat group as a sample,our study is to explore the temporal characteristics of social user group interaction behavior,and to provide the reference for directional information delivery and public opinion guidance.[Method/process] Wechat user’ sharing and reading behavior records are collected through self-built APP platform,and the connection tendency of interaction users and time characteristics of interaction behaviors are quantitatively analyzed with the help of human dynamics.[Result/conclusion]The time interval distribution of user sharing behavior in Wechat group deviates from the exponential distribution obviously and has a strong burstiness effect;the users in the group tend to have heterogenous connections,but most users appear only as bystanders;the articles titled with emotional induction and benefit guidance can arouse more users’ interaction willingness.[Limitations]The collected data was not comprehensive which might generate some biased results.