基于语义扩展的微博突发事件检测方法及装置

• Format: Patent
• Language: Chinese
• Published: 08.11.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

本发明提供一种基于语义扩展的微博突发事件检测方法及装置，本发明提供的基于语义扩展的微博突发事件检测方法包括：获取与待检测的微博突发事件对应的第一关键词；根据第一关键词和第一关键词关联的词向量文件，得到微博突发事件；其中，第一关键词关联的词向量文件是采用word2vec方法对训练集中的训练词语进行训练得到的，词向量文件包括第一关键词与其它训练词语之间的语义相似度。本发明的基于语义扩展的微博突发事件检测方法及装置，通过考虑文本的语义信息来扩展事件的关键词，并通过扩展后的关键词进行微博突发事件的检测，可以准确的检测微博突发事件，避免了将同一事件分成两个微博事件，并可以进行在线检测。 The invention provides a semantic extension-based microblog emergency detection method and device. The semantic extension-based microblog emergency detection method provided by the invention comprises the following steps of: obtaining a first keyword corresponding to a to-be-detected microblog emergency; AND obtaining a microblog emergency according to the first keyword and a word vector file associated with the first keyword, wherein the word vector file associated with the first keyword is obtained through carrying out training on training words in a training set by adoption of a word2vec method, and the word vector file comprises semantic similarity between the first keyword and other training words. According to the se