Privacy-preserving collaborative filtering based on horizontally partitioned dataset

Abstract

Nowadays, recommender systems have been increasingly used by companies to improve their services. Such systems are employed by companies in order to satisfy their existing customers and attract new ones. However, many small or medium companies do not possess adequate customer data to generate satisfactory recommendations. To solve this problem, we propose that the companies should generate recommendations based on a joint set of customer data. For this purpose, we present a privacy-preserving collaborative filtering algorithm, which allows one company to generate recommendations based on its own customer data and the customer data from other companies. The security property is based on rigorous cryptographic techniques, and guarantees that no company will leak its customer data to others. In practice, such a guarantee not only protects companies' business incentives but also makes the operation compliant with privacy regulations. To obtain precise performance figures, we implement a prototype of the proposed solution in C++. The experimental results show that the proposed solution achieves significant accuracy difference in the generated recommendations.

Privacy-Preserving Context-Aware Recommender Systems: Analysis and New Solutions

“…For example, the solutions from [2, 8, 9, 12-15, 20, 24, 25, 28, 32, 36] fall into this category. More specifically, the solutions from [2,8,9,12,15,20,32] focus on distributed setting where every individual user is expected to participate in the recommendation computation, while those from [13,14,24,25,28,36] focus on partitioned dataset, where several organizations wish to compute recommendations for their own users by joining their private dataset. These solutions typically assume semi-honest attackers and apply existing cryptographic primitives to secure the procedures in standard recommender protocols.…”

Privacy-Aware Detection of Shilling Profiles on Arbitrarily Distributed Recommender Systems

“…However, corporate privacy is the biggest obstacle to such cooperation, since collected data is a valuable and confidential asset of a company. Besides confidentiality concerns based on legal regulations, companies might hesitate to collaborate due to financial fears [15], [23]. Various PPDCF schemes are proposed enabling data holders' collaboration without jeopardizing their privacy.…”

Privacy-Preserving Planarity Testing of Distributed Graphs

“…The strict notion of perfect privacy-preservation is sometimes relaxed by allowing some leakage of information, if such a relaxation enables a more efficient computation and if the leakage of information is characterized (in order to decide, in any given application setting, whether the gain in efficiency justifies the reduction in privacy-preservation). There are many examples of studies that relax perfect privacy in order to allow practical solutions, from various domains such as distributed association rule mining [4,5], anonymization of distributed datasets [6][7][8], collaborative filtering [9,10], distributed graph mining [11], and distributed constraint optimization problems [12][13][14].…”