Abstract:

As for the problems that the input query are usually ambiguous and concise in the current information retrieval, this paper presents a personalized query recommendation method based on the TFIQF model and graph clustering. It provides query recommendation for a giving query, helps the users to amend query and retrieves the information needed. Simultaneously, it will achieve the goal of providing personalized query suggestions by capturing different user preferences. The experimental results show that the method gives personalized query suggestions, provides the potential resources that the user is interested in and obtains high precision which improves the quality of personalized retrieval systems.

Key words: personalized search;query recommendation;TFIQF model;graph clustering