This paper designs a personalized information search system(named PISS) applied to online shopping. According to users’ searching issue and information database, PISS provides some initial interest merchandise. After analyzing the performance of the merchandise, the system records the result and returns it to the database. PISS can be used in finding consumers’ needs and mining the data of connotative expression. The innovation of this paper is providing solution to related actions of users and designing the interesting model.