The Discovery And Analysis Of Micro-blog's Public Opinion

With the rapid development of information technology,the Internet has been widely spread,and it has a far-reaching impact on society.As an important product of the Internet age,micro-blog has become an integral part of the lives of many Internet users.Micro-blog is a broadcast social networking platform for sharing short,real-time information.On the micro-blog platform,users can describe what's going on around them,express their views on something,and even promote their products.While meeting the needs of social networking,micro-blog has shown an important position in the network of public opinion.How to dig and analyze micro-blog public opinion in the hundreds of millions of micro-blog texts is becoming more and more important.At present,the research on micro-blog's public opinion is still in its infancy,and there are still a lot of work worth studying.The main research work of this paper is carried out in the following three aspects.1.Micro-blog data crawling,screening and pretreatment.This paper studies micro-blog data acquisition scheme based on micro-blog API and crawler.In this paper,two micro-blog data acquisition schemes are used to collect public micro-blog data and micro-blog data.Micro-blog data is screened from micro-blog users and micro-blog content.Moreover,micro-blog text is processed by word segmentation and Stop word removal,feature selection and vector representation.2.A micro-blog public opinion discovery algorithm is proposed.The existing micro-blog hotspot discovery algorithms are analyzed,the advantages and disadvantages of each hotspot discovery algorithm are studied,and a HEA hotspot discovery algorithm is proposed.Aiming at the problem that single text clustering algorithm has poor effect on micro-blog text clustering,a hybrid clustering algorithm named HKSK is proposed.Combining the HKSK algorithm with the HEA algorithm,a hot spot detection algorithm of HKSK-HEA is proposed.The experiment proves that the algorithm can effectively detect the micro-blog hotspot.3.A micro-blog sentiment analysis algorithm is proposed.The current emotion analysis algorithms are studied.The existing emotional dictionaries are sorted out and thebasic emotion dictionary is constructed.In addition,an emotion correction dictionary is constructed according to the semantic features of text.Considering the network properties of micro-blog text,a dictionary of emoticons is constructed.And the application of Naive Bayes classifierier in emotion classification is studied.Aiming at the problem of low generalization ability of a single emotion classification algorithm,a method of fusing the three is proposed,and an emotion Naive Bayes classifierier is designed.Experiments show that the emotion Naive Bayes classifierier can improve the performance of micro-blog sentiment analysis algorithm.