论文部分内容阅读
以微博、微信、社交网络等为代表的微媒体文本传播方式迅速成为大众关注的焦点。本文首先概括了微媒体文本热点事件传播的特点;接着利用数据挖掘的方法对微媒体文本信息流进行预处理,形成高维向量空间;而后根据其时序性特点在每个时间区域内进行聚类,抽取出大类信息,从而形成热点事件;最后根据热点事件的生命周期和强度变化趋势,分析出热点事件的发展趋势。
Microblogging, WeChat, social networking, as the representative of the micro-media text communication quickly became the focus of public attention. This paper first summarizes the characteristics of the propagation of hot events in micro-media texts, and then uses the data mining method to preprocess the micro-media textual information streams to form high-dimensional vector spaces. Then, clustering is performed in each time domain according to its temporal characteristics , Extract a large class of information, thus forming a hot event; Finally, according to the life cycle of hot events and trends in the intensity of change, analysis of hot trends in the development trend.