Design And Implementation Of Case-Based Learning System For Optimizing Results From Search Engine

Search engine is main tool for searching information on internet, and it facilitate the people in information retrieval, but, it is different for users to find what they really want with the existing search engine, consisting in too many record the search engine returned. Furthermore, users can not describe the information they need simply with the several simple words. Give a query, the search engine usually returns thousands upon thousands of text results, which are dynamic and brief, the most parts of them are irrelevant to specific user, so the users have to browse through a long list to get what they want. As a result, the questions like "information overload" and " information losing" appeared in information retrieval. How to improve the search engine's precision is a primary problem in development of search engine, and how to infer the user's query purpose in order to achieve the intelligent search is the development direction of search engine in the future.Besides query words tends have different meaning and different users have various background, interests and usage intents. At present, for the specific query word, the search engine give the same result list between the different users. With the hoping that the results can be consistent with their own wishes, a variety of improved search engine appear, including search engine based on the user's personality dictionary, search engine based on clustering technology, subject-oriented search engine, and so on. They promote the progress in search engine to a certain extent. In principles, it can be regard as the process of results with the actual technology, including filtering of results, clustering, classification, and so on.Case-Based Learning is the more mature branches, the basic idea is to obtain a general rule that can be used to exclude every negative example, and which include all the positive examples, by means of induction and conclusion of the set of positive and negative examples of conception given, which is also known as the concept of access. This article is based on track of user's behavior, it divide the web page visited into negative and positive set, get the rule that represent user's purpose through the application of corresponding arithmetic, thereby realize the filter of web page, and the result include the most relative record and exclude irrelative record, effectively improve the accuracy of search engine, and provide users with high-quality, high _correlation results.Based on the analysis of the general search engine and personalized search engine, this article presents a strategy to improve search engine in many aspect, which based on track user's behavior in visiting web page, mine the information of summary of returned web page, and deduce the aim of the user, in the end, optimize the search result. It can remove garbage information in search result, return to the user a more satisfactory results. Finally, in this article realize the search engine optimization system(SEO), the system performed well in test.