Research On Text Information Filtering Technologies Based On Semantic Orientation Analysis

Language is the main carrier of information. In the people's everyday work, most of the information is expressed, recorded, delivered and exchanged through language. Therefore, with the popularization and application of the computers and Internet, the data and information processing have developed to knowledge processing, the request for the language processing degree is becoming more deeply and widely. Since most of the data are massive in scale and diverse in subject areas, they make the information acquisition bottleneck more severe, thus greatly challenge the processing systems in speed, precision and robustness.Most of the research on text filtering is about thematic text filtering, but with the fast development of the Internet, tendency text filter is giving more effects on text information security. Text filtering technology based on statistics usually is ineffective when it deals with polarity text. The method overlooks the semantic restriction of text, so it isn't good for identifying polarity information.This dissertation briefly describes the background of text filtering any discusses the stematical relationship of text filtering and text categorization, machine learning, etc. By analyzing how situation affects text understanding, the thesis sets up relations between texts and situation models and relativity of text features.Finally, we come up with a planning project of the information filtering based on the tendency text filtering technology, and describe it from text presentation to the function of weights in detail. Of course, the precision of filtering is not ideal, so the next contents of this subject are summarized systematically and some one's own views are also presented.