Privacy-preserving association rule mining based on electronic medical system

Abstract

Privacy protection during collaborative distributed association rule mining is an important research, which has been widely used in market prediction, medical research and other fields. In medical research, Domadiya et al. (Sadhana 43(8):127, 2018) focused on mining association rules from horizontally distributed healthcare data to diagnose heart disease. They claimed they proposed a more effective privacy-preserving distributed association rule mining (PPDARM) scheme. However, a serious security scrutiny of the scheme is performed, and we find it vulnerable to protect the support of the itemsets from any electronic health record (EHR) system, which is the most important parameter Domadiya et al. tried to protect. In this paper, we first present the cryptanalysis of the PPDARM scheme proposed by Domadiya et al. as well as some revised performance analyses. Then a new PPDARM scheme with less interactions is proposed to avert the shortcomings of Domadiya et al., using the homomorphic properties of the distributed Paillier cryptosystem to accomplish the cooperative computation. Our scheme allows the directed authority (miner) to obtain the final results rather than all cooperative EHR systems, in case of semi-honest but pseudo EHR systems. Moreover, security analysis and performance evaluation demonstrate our proposal is efficient and feasible.

LRFM- based association rule mining for dentistry services patterns identification (Case study: a dental center in Iran)

“…They identi ed 259 different rules with more than 95% con dence level. (Xu et al, 2022) tried to extract various for privacy-preserving using association rule mining algorithms based on electronic health record (EHR) system. (Han & Zhang, 2022) also tried to identify the multi-morbidity patterns for different ages and genders.…”

Privacy-enhancing data aggregation and data analytics in wireless networks for a large class of distributed queries

“…On the other hand, the protection of privacy is also crucial for successful applications of novel cyber technologies [5][6][7]. For example, privacy-preserving techniques have been investigated for data publishing [8], service selection [9], and trend surface analysis [10].…”