Privacy Preserving Association Rule Mining without Trusted Party for Horizontally Partitioned Databases

Many data mining techniques are available to explore useful hidden information from large databases. Among these, association rule mining has wide applications to discover interesting relationships amongattributes. The issue of privacy arises when the data is distributed among multiple sites and no site owner wishes to provide their private data to other sites but they are interested to know the global resultsobtained from mining process. In this paper a new model is proposed which utilizes hash based secure sum cryptography technique when no site can be treated as trusted party to find global association rules for horizontally partitioned databases.