In this paper, we study the problem of outsourcing the association rule mining task within a corporate privacy-preserving framework. A substantial body of work has been done on privacy-preserving data mining in a variety of contexts. A common characteristic of most of the previously studied frameworks is that the patterns mined from the data (which may be distorted, encrypted, anonymized, or otherwise transformed) are intended to be shared with parties other than the data owner. The key distinction between such bodies of work and our problem is that, in the latter, both the underlying data and the mined results are not intended for sharing and must remain private to the the data owner. We adopt a conservative frequency-based attack model in which the server knows the exact set of items in the owner’s data and additionally, it also knows the exact support of every item in the original data. The early works on defending against the frequency-based attack in the data mining outsourcing scenario. They introduced the idea of using fake items to defend against the frequency-based attack. The proposed encryption/decryption scheme is a viable solution for privacy-preserving pattern mining over outsourced TDB, provided that a correct and efficient implementation exists. On the efficiency side, it is not practical to store the support.