Research On Text Clustering For Search Engine

Under the popularity of network, people come into a comprehensive e-commerce era.Clothing, books, electronics, appliances, and even food are beginning overwhelminglyappear on the Internet. Facing the situation an array of goods that appear on the network,how to make their products stand out, how to enable users to select the most suitable fortheir own goods, has become a battleground for each merchant. Which means, not onlyimprove the quality of their products, but also concerned about the user’s personalizedservice, has become the research’s intentions. Research the true intentions of the user, get amore accurate search results has become a research hotspots.With the widespread of internet, search has become an indispensable behavior inpeople’s lives. Search engines not only search from traditional search engines such asBaidu, Google but also the broader-based B2B, B2C, C2C search applications environment.In order to cater the user’s search needs, people began to focus on the pre-handle searchresults by text clustering search. The article not only concerned about how to get a gooduser experience through text clustering, but also based on the premise of current rapiddevelopment of information condition, e-commerce has spread to people’s vision. Thispaper describes the current search engineers shortage and text clustering algorithm, inorder to better meet user’s personalized service, proposed KM-BKW clustering algorithmprototype system model, track user browsing behavior, analyze user interest informationthrough k-means clustering algorithm to return to the search engine interface, allowingusers to get a new experience.The main contents of the article are as follows: The first part of a review of theextensive literature on the origins of clustering search engine, introduced the development,and search engine clustering classification, the inadequacies of the current search enginesclustering summarized, explained. The second part have a detailed introduction to the textclustering knowledge. The third part describes the search engine information processing approach. The forth part summarizes internet search engine application characteristics,innovative proposed KM-BKW clustering algorithm. The fifth part builds a prototypesystem based on the KM-BKW clustering algorithm, using KM-BKW algorithm to processdata, pay attention to the personalized service users, get the accuracy of experimentalresults. The sixth part is a review of this text, found the future development of searchengine results clustering problems and prospect.