Distributed K-Anonymous Location Privacy Protection Algorithm Based on Interest Points and User Social Behavior

Abstract

Location-based services have become an important part of our daily lives, and while users enjoy convenient Internet services, they also face the risk of privacy leakage. K-anonymity is a widely used method to protect location privacy, but most existing K-anonymity location privacy protection schemes use virtual locations to construct anonymity zones, which have the problem of being vulnerable to attackers through background knowledge, while the improved collaborative K-anonymity scheme does not sufficiently consider whether collaborating users share similar attributes. We propose a distributed K-anonymity location privacy-preserving algorithm based on interest points and user social behaviors to solve these problems in existing K-anonymity schemes. The method determines the similarity of users by their interest points and social behaviors and then selects users with high similarity to build an anonymous set of collaborative users. Finally, to ensure the relatively uniform distribution of collaborative users, a homogenization algorithm is used to make the anonymous location points as dispersed as possible. The experimental results showed that our algorithm can effectively resist background attacks, and the uniformly distributed anonymous location points can achieve higher-quality anonymous regions.