Research On The Application Of Text Mining Based On Content Analysis In Network Marketing Strategy

The rise of Web2.0provides customers with a platform to express their views,a growing number of network messages and product reviews emerge in out our site.These comments not only help customers understand the reputation of the productsand services but also can be used as a feedback mechanism to help manufacturersand vendors to improve product quality, improve service attitude, and improvemarketing strategies to gain competitive. With the rapid growth of the networkreviews,the information contents which have useful contents have been buried in alot of useless, irrelevant information, and obtainings the knowledge like a needle ina haystack. How to use a certain method to mine the features of product,obtain thecharacteristics, understand the level of satisfaction of users, and help sellersimprove the marketing strategy have become a hot topic in today’s research.Product reviews text mining in Web2.0environment is the focus of thisdissertation. This dissertation describes the existing methods and introduce theimprovements based on the existing methods. At the same time, combined withcontent analysis method on the comment text,find the characteristics which controllthe sales importantly,and improve the marketing strategies for sellers. The mainwork of the parper have include five aspects:(1)On the basis of reading the relevant information of text mining, haveunderstanded the knowledge of methods and techniques of text mining, andanalysis the new features of the comment text in web2.0environment. Improve theoriginal text mining method and improve the mining results.(2)Study the psychological and behavioral characteristics of consumers in theweb2.0environment, obtain the product reviews information on the web page andanalysis of the text of comments on the marketing model, pointing out that thecomment text seller marketing strategy developed a supporting role.(3)Finish the comment text mining experimental design, choose thehighest-selling products in Taobao, observate test and analysis the goodsindependently in the Matlab simulation system.(4) This article contains two points of technological innovation: first, the wordof features extraction method to improve the K-center point algorithm combinedwith the characteristics of content analysis, divided by the coder training and coding indicators, to identify newquantitative indicators, unstructured qualitativequantitative analysis of comment text.