Web-based Public Opinion Analysis And Forecast Trend Analysis Text Orientation

With the popularity of Internet, the network public opinion has graduallybecome the power with important influence to the social public opinion and to a greatextent it leads the public understanding of some public opinion events, therefore, thepublic opinion analysis technology arises at the historic moment, it can be used by thegovernment or enterprise to understand and master the development status of publicopinion, and to provide scientific analysis, so as to improve the decision-making levelof the government or enterprise and work efficiency.This paper begins with the main theoretical knowledge of network publicopinion analysis and key technology involved, on the basis of text tendency analysiswe can put forward a more reasonable public opinion analysis method which changesthe previous public opinion analysis solely from the perspective of quantity analysismethod or substance analysis, but combine two ways as one. Through the method ofquantification weighting with the public opinion in terms of emotional tendency wecan do a further research so that we can get better understanding of public opinion, soit embodies the mutual combination of qualitative analysis and quantitative analysis.On the basis of the theory we built a complete framework can be used to analysisnetwork public opinion and finished a semi-automation system, with a representativeexperiment to show the feasible and scientific of this theory, besides, The work doneby this paper also includes:(1) We constructed a emotional lexicon in the field of network public opinion analysis,thus lay a foundation of network public opinion analysis based on Chinese text andcan be a reference to others.(2) After the network public opinion analysis we quantified the result, this methodreflects the idea of qualitative analysis in quantitative analysis and so that we cangrasp the overall situation of network public opinion.(3) On the basis of public opinion analysis which has been carried out we forecastedthe development of public opinion.(4) On the basis of public opinion analysis and the opinion of the paper we improvedthe algorithm of the emotional tendency.