To improve the retrieval efficiency and performance of the large scale information retrieval systems, analyzed existing replacement algorithm for WEB caching, due to the diversity of the WEB traffic pattern, the traditional algorithms for cache updating can not be used in WEB environment effectively. In this paper, with click-through data analysis, a inverted file replacement algorithm for WEB caching is proposed. The analytic result shows that the click-through data for the cache updating algorithms is how the algorithm suits the WEB traffic pattern properly. Based on the poisson arrival model, a new cache policy, inverted file replacement algorithm, is proposed. The trace driven simulation shows that the retrieval algorithm under the new organization of the inverted file can decrease its execution time significantly and the performance of the inverted file replacement algorithms is better than that of the existing algorithms proposed in the literature.