Research On Personalized Meta Search Results Merging In Information Retrieval

How people can find the information they are interested in from the Internet becomes the most important problem. The Meta-Search Engine provides more comprehensive information of searched results, but it does not improve the search precision. For settling this problem, an algorithm of meta search results merging based on user profile is proposed.In this paper, firstly, information retrieval and Meta-Search Engine were researched, including the key tech in meta search results merging. Secondly the user profile model and an algorithm on user profile vectors mapping from user query were researched. Meta search results merging algorithm proposed in this paper included lemma and user profile based on Abstract algorithm which is used to estimate the similarity between the results and user query, and realized user personality. In the end, a prototype has been developed for providing personality information retrieval based on user profile, and a lot of experiences comparing this algorithm with single search engine and other results merging algorithms were made. This algorithm was proved to ensure the full-scale and accuracy of the search results, improve the effect and efficiency of the searches and satisfy user personality requirements.