摘要
【目的/意义】随着社交网络的普及与快速发展,人们越来越多地依赖于网络聊天工具进行交流,针对QQ群组聊天信息过载用户无法从聊天记录中快速获取所需信息的问题,本文开展了聊天热点主题提取和QQ群组用户聊天行为分析的研究。【方法/过程】采集了一个技术类QQ群的聊天数据,利用Gibbs算法和LDA模型提取群组聊天数据中的主题并对其进行分析。【结果/结论】发现群组的主题可以分为三类:技术类、生活类和综合类。其中,技术类话题讨论的高峰集中在工作时间,没有继承性;大家普遍关心生活类话题,该话题有继承性。由于群组聊天的即时性、交互性和网络领袖的影响,一个时间段内群中只有一个热点主题。该研究结果可为群组聊天行为和热点分析提供参考。
[Purpose/significance] With the popularity and rapid development of social network, people are increasingly relying on the online chatting tools to communicate. In light of the problem that user can＇ t extract useful information from the overload chat record quickly, this paper adopts LDA model to extract the hotspot information from QQ group chats; besides, we also analyze the chat behavior. [Method/process] Firstly We collect the chat data of a QQ online technical group, then extract the topic through Gibbs algorithm and LDA model, and analyze chat data. [Result/conclusion] We find out that the topic of this group can be classified into three categories： technical, life-related, and comprehensive. Besides, the peak time of the discussion of the technical topics is focused on the working time, and there is no succession of technical topics; the life-related topic is mostly concerned by people, and there is succession of life-releted topics. Apart from this, there is only one hotspot topic in a period of time in a group due to the instantaneity, interaetivity and the influence of the network leader. Our finding can provide a reference for researching on group chats behavior and hotspot analysis.
出处
《情报科学》
CSSCI
北大核心
2017年第12期45-49,共5页
Information Science
基金
国家社科基金(17BSH135)
