Abstract: For effectively mining useful rules, combining the individual databases into a single logical database is not appropriate in distributed environment In this paper a concept of maximum weighted frequent itemsets is introduced, a novel model for measuring database similarity is given, and an algorithm based on maximum weighted frequent itemsets for measuring database similarity is presented Once similar databases are clustered, each cluster can be independently mined to generate the appropriate rules for a given cluster In real applications, the new model provides an effective framework for data preparation of data mining in distributed environment, and hence it has real importance