Research On The Evolutionary Search Algorithms In The WEB Based On Learning

With the development of internet and the technology of the web information, it is impossible to find out the useful information in the information ocean. In other words, it is the same difficult as looking for a needle in the ocean. However, the search engine happens to solve the problem (it can provide the information retrieve for users). For the moment, the technology of the search engine is becoming the focus of the industry world and academy.The existing catalog search engine collects information by the labor or half-automatism. The collected information will be input in the certain catalog after they have been examined by the editors and the information abstract has been finished. Here, there is a plenty of inferior use of the labor and time. What is more, when we retrieve the information from the www by the key words, the result is a list of linear documents. When the result is reached to tens of pages, it will take us much energy to retrieve the information, which we are bored with.In this paper, we praise the point that it is an efficient way to solve the intelligent search in the search engine by the analysis of the clustering technology and ant algorithm. First, we have expatiated the working principle, performance parameters and major technologies. Farther, we have analyzed the shortcomings of the existing catalog search engine and introduced the clustering analysis and the ant algorithm; on the basis of this, we discussed the possibility and necessity of the connection between them, which avoids the local optimization of the clustering analysis to a degree. In the end, we appraise the idea that we deal with the information data by the data structure of the binary tree, m-branch tree and tree established by the ant algorithm, which can improve the efficiency of the search engine. In the paper, we have built the binary tree, m-branch tree and tree algorithms based on the test data. What is more, the cluster results have improved obviously after we have developed these algorithms, namely, we have thought over the density information of the data during the course.In the paper, we have carried out the corresponding comparison and have gained a series of important conclusions.