A lambic : a privacy-preserving recommender system for electronic commerce

摘要: Recommender systems enable merchants to assist customers in finding products that best satisfy their needs. Unfortunately, current recommender suffer from various privacy-protection vulnerabilities. Customers should be able keep private personal information, including buying preferences, and they not tracked against will. The commercial interests of also protected by allowing them make accurate recommendations without revealing legitimately compiled valuable information third parties. We introduce a theoretical approach for system called Alambic, which achieves the above objectives hybrid combines content-based, demographic collaborative filtering techniques. Our splits customer data between merchant semi-trusted party, so neither can derive sensitive share alone. Therefore, could only subverted coalition these two