Web News Opinion Orientation Analysis Based On Semantic Understanding

With the dramatic delevopment of web tecnology, the number of freely available online reviews is increasing at a high speed. A significant number of websites, blogs and forums allow users to post reviews for various News Events. In the past few years, mining the opinions expressed in web reviews attracts extensive researches. Based on a collection of customer reviews, the task of opinion mining is to extract customers’ opinions and predict the sentiment orientation. Opinion mining actually identifies the author’s viewpoint on an event, rather than simply identifying the event itself.To this end, this paper, based on senmantic understanding presents analyzing opinions orientation of Web News.In the theory of the system, the overall framework, the major functional modules, and from the Internet through a large number of test data show that the framework is feasible. In this work, we propose to analyze the opinion orientation of the topics and judge the opinion polarity.Major work: Firstly, key sentence in each paragraph is extracted based on the word segmentation and dictionary of polarity words. Secondly we calculate the tendency value of key sentences with the standpoint. Thirdly, combined with the position of key sentence and paragraph, key weights are calculated. Finally, tendency analysis of web news was realized by calculating tendency values and weights of key sentences. Experimental results on mid-scale web pages show that the method is effective. This public opinion monitoring system can help us to discover opinions of hot topics, especally the web news with extreme opinions. It can realize the topic tendency tracking task as well and maintain web content security.