A non-word-segmented burst topic detection method for microblog

Abstract

The invention relates to a non-word segmentation emergent topic detection method for a microblog, which utilizes a computer technology to assist the intelligent analysis or public opinion of network information. The method comprises the following steps of pretreating corpus, and establishing a dynamic microblog detection window; segmenting the content of the microblog into single Chinese characters, and establishing a dictionary; calculating a collection of emergent feature characters; calculating an emergent topic consisting of feature characters; generating meaningful words or strings, and forming the emergent topic formed by the words or strings. The detection method has the advantages that the Chinese word segmentation is not needed for the emergent topic of the microblog, the content of the Chinese microblog information does not need to be subjected to Chinese word segmentation in advance, and the Chinese characters, English words, pictures, videos, external links and the like are respectively used as single entities; finally, the Chinese characters in the entity of the emergent feature form the words, so the integral property is improved, and the recall rate of new words and colloquial strings is improved.

Description

technical field [0001] The invention relates to a microblog-oriented non-word-segmented burst topic detection method using computer technology to assist intelligent analysis of network information or public opinion. Background technique [0002] With the rise of the mobile Internet, microblogs such as Sina and Tencent have risen rapidly in my country. The number of registered users of Sina Weibo has exceeded 200 million, and the number of registered users of Tencent Weibo has exceeded 160 million. The number of microblog messages generated in my country's microblog network exceeds 300 million every day, and the microblog platform has become one of the main channels for people to obtain news and information in their daily lives. Due to the push mechanism of Weibo, the news of Weibo spreads rapidly in the network, which produces a huge spread influence. [0003] While Weibo provides people with information, it also increases the difficulty of social management. Events in th...

AI Technical Summary

Problems solved by technology

The Weibo network contains a large number of social and colloquial words, and it is difficult to detect sudden topics induced by new words or strings based on word segmentation methods

Images