In modern times, it is convenient for the development of information and communication, and has enjoyed various merits. However, the times have changed, personal diversity is increasingly emphasized. In keeping pace with the increasingly accelerated personal era in these developments, we research and analyze the existing standardized book recommendation system, select directly several categories in the past, and each time a user Investing a lot of time, solving searching trouble, grasping the advantages and disadvantages of the system, maintaining the existing system, even if the user does not invest time directly, personal preference and We propose a "personalized book recommendation system" recommended based on purchase/loan information. This system consists of server based on information management, book information management, book purchase / loan management, analysis, administrator, book recommender, DB administrator. Members via each administrator are booked, purchased, loan information is stored and managed, and based on the statistical information using the analysis manager, the recommender recommends books that meet members' personal preferences It will come to you. Based on the system studied in this thesis, applying the information profile of the user (who can identify individuals, interests, evaluations, etc.) can be utilized in various fields such as movies, music, research themes, products, etc., From the viewpoint, it is expected that it can contribute to the revitalization of related markets.