【摘要】 基于实现网络谣言自动识别的目的,从地域、时间和传播形式3个维度分析了收集到的网络谣言基本情况。网络谣言以文本传播形式为主,而且在文本句式上有一定的共通点和相似之处。本文采用了五类网络谣言文本句式特征分析方法,结合LanguageTool构建了一系列基于XML的网络谣言句式匹配规则。通过对收集到的网络谣言实验测试,得出此方法能够实现网络谣言的自动识别和监测,可以减少50%以上的人工识别工作量的结论。更多还原

【Abstract】 In order to realize online rumors automatic identification,it introduces the collected online rumors in the three dimensions of location,time and spreading media. The main spreading media is text based on analysis,with common and similar text syntactical structure features. Five text syntactical structure features are summarized,based on which a series of online rumor text syntactical structure XML rules are constructed. Tests are performed upon the collected online rumors,which shows that it can realize the automatic identifying and monitoring the online rumors,with half of the manual work-load reduced.更多还原