Abstract: User-oriented materialized views are able to greatly improve OLAP query performance for users. However, the available methods for cache management are not able to deal with the issue of dynamic view selection, since they do not take into account the data access pattern of OLAP queries of specific users. In this paper, the concepts of view path and view tree are proposed to organize the views. Also, a method called reverse path growing is proposed to quickly compute view path for a newly-arrived query, so as to greatly reduce query response time. Furthermore, an effective view replacement method based on reserved view path is designed to better deal with the issue of dynamic adjusting of view tree. Extensive experiments show that the proposed method can achieve better performance than those previous ones.