Web Information Systems and Mining

Abstract

How to get the Internet public opinion timely and comprehensively is a key problem for local governments. The method of keyword matching lack of analysis and statistics on semantic, reduced the precision and recall. In this paper, the detection algorithm of semantic expansion is proposed by utilizing Integrated E-Government Thesaurus (Category Table) to construct domain ontology and through expanding the area knowledge. The experimental results show that the precision rate and recall rate of public opinion detection have all been improved obviously. © Springer-Verlag Berlin Heidelberg 2012.