Micro-blog Query Expansion Based on Lexical Time Distribution

摘要

This paper presents a method of query expansion based on lexical time distribution for micro-blog search. Similarity of distribution is used to measure the correlation between the extended words and the query words. A query model based on the time distribution of the words is established. The query expansion method is unable to accurately estimate the expansion words due to the short micro-blog content. The real graph on the evaluation data is retrieved by the TREC2011 and TREC2012 micro-blog. The results show that the model can effectively improve the performance of micro-blog search extension vocabulary query based on time distribution, not only significantly outperforms the classical content based query expansion model, which is better than other time using the method of query expansion.