The Research Of Query Expansion Technology On Short Texts

In recent years, microblog has become a new network application which combines the social networking sites and the mass media. It plays an importance role in spreading breaking news and profoundly impacts on public opinion towards. Facing tens of millions of real-time updating microblogs, users need a convenient and effective way to access information. Therefore, the requirement of microblog retrieval is becoming more. As a key technology in the field of information retrieval, query expansion is vital to optimize retrieved results. The main contents of this thesis include:Firstly, we propose a Bayes-LDA based modeling method on microblog. The model can guarantee the quality and completeness of the modeling on short texts such as microblogs. Secondly, we design a query expansion algorithm based on topic model. Its core thought is to apply the modeling results of Bayes-LDA to the generation of expansion features and the re-ranking of search results. Lastly, a real-time processing system for massive data is introduced. The thesis describes the author’s modules, i.e. the stream processing framework based on Map-Reduce and the storage solution combined features of database and search engine.