Validating the Performance of Personalization Techniques in Search Engine

User profiling is an important and basic component in personalized search engine. Search engines respond to a user's query by using the bag-of-words model, which matches keyword between the query and web documents but ignore contexts and users' preferences. Personalized search greatly improves the search results as of the results provided by the search engine without personalization. In this paper, the performance of personalized search based on content analysis and personalized search based on user group have been evaluated. In personalized search based on content analysis the contents are traced by finding the user's browsed documents and search history, which reduce the users search time. In user profile only user preference alone is taken into consideration. The experimental results show that the personalized search based on user group method having higher precision and recall rate than the content analysis method.