Accuracy Enhancement of Mining Association Rules
Priyadarsan Patra · 2014
I. Introduction An vast quantity of privately owned records that express individuals, interests, activities, and demographics. The records often include sensitive data and may violate the privacy of the users if published. The information is suitable for very important resource for many systems and corporations that may improve their services and performance by remind novel and potentially useful data mining models. One of the common practice for releasing such confidential data without violating privacy and apply some regulations and policies for the data usage. These type of regulations usually entail data distortion operations such as generalization . The challenge with this approach is the data leakage can still occur and the data and the resulting data mining models may become nearly useless after excessive distortion[8]. The upcoming research field of Privacy Preserving Data Publishing (PPDP) is targeting this challenge [8]. PPDP also aims at developing techniques that allow publishing data while minimizing distortion for maintaining utility on one hand and ensuring that privacy is preserved on the other. In this paper we present a new privacy-preserving data publishing(PPDP) method, which is shown to preserve the