Abstract: In order to apply frequent itemsets mining and query expansion technique to information retrieval， a novel informa-tion retrieval system is introduced based on fusion of frequent itemsets mining and query expansion. To set forth its designing conception， the general structure of the system model and the function of each module are expatiated， and searching algorithm of the system is also expounded， a prototype system is realized. The results of the experiment show that the proposed retrieval system can effectively improve and enhance the performance of information retrieval.

Key words: information retrieval, frequent itemsets, query expansion, model