Short Text Classification Based on Rule and Information Entropy

With the development of computer technology and the progress of the means of information dissemination, microblog, social networking and other information in the form of short text develop rapidly. Mining useful information and filter harmful information from short text become particularly important and urgent. In recent years, technology of classification and filter from lengthy document has made a considerable development, but for short text classification, traditional text classification techniques are no longer applicable due to the small number of words, short length of text and less information contained. In this paper, taking the topic classification of microblogs of Sina as application background, short text classification algorithm for rule-based information entropy is presented, as an effective algorithm to improve the effect of short text classification and meet the needs of it through experimental analysis.