Fake-locus-based location privacy protection method for use in recommendation system

Abstract

The invention discloses a fake-locus-based location privacy protection method for use in a recommendation system. According to the technical scheme, the method comprises the following steps that: 1, a communication system framework consisting of a user, a public location database and a recommendation service operator is built; 2, the user caches an actual location locus, and converts the actual location locus into a corresponding semantic locus; 3, the user builds a location class hierarchy tree according to a semantic class of location data in the public location database, and performs clustering according to the similarity of the location data; 4, the user generates a fake locus through a Viterbi algorithm; and 5, the user uploads the fake locus to the recommendation service operator in order to protect the location privacy of the user. According to the method, a semantic class of an actual location of the user is taken as an input of the Viterbi algorithm to generate the fake locus, and the actual location of the user is not involved in computation, so that the service operator cannot speculate the location of the user; the location privacy of the user is protected; and the service quality of the recommendation system is enhanced at the same time. The method can be applied to a mobile social network for making recommendations according to interests of the user.

Description

technical field [0001] The invention belongs to the technical field of wireless networks, and relates to the protection of location privacy and a mobile social network for user interest recommendation. Background technique [0002] Social network is a form that reflects social communication groups that naturally develops with the application of the Internet such as Email, BBS, blog, and Weibo. platform. With the popularity of various mobile devices such as smart phones and vehicle-mounted mobile terminals, and the application of sensor network technology, it is becoming more and more popular to use mobile terminal devices to access social networks, and mobile social networks have gradually become mainstream. According to the latest report released by Nielsen in February 2013, more than half of social network users worldwide access social networks through mobile devices such as mobile phones. Among all the applications of mobile social networks, location-based service (LBS)...

AI Technical Summary

Problems solved by technology

However, the inaccuracy of location information will lead to the reduction of recommended service quality. The degree of location fuzzy processing must meet the requirements of service quality. protection of

[0008] To sum up, using the above two methods will still lead to location leakage to a certain extent, and the quality of service obtained is also limited

[0009] In addition, in the above two types of methods, k-anonymity achieves location privacy protection by hiding the user's real location in some false locations, and the obfuscation-based method of location offset is to make a certain geographical difference to the original location. Deviate or generalize the location so that the server or attacker cannot determine the user's true location

However, current k-anonymity and obfuscation-based privacy protection basically do not consider the semantic properties of location data, and the generated fake locations may be semantically completely unrelated to real locations

Therefore, applying the above two location privacy protection techniques to a location recommendation system will lead to a decrease in service quality

Images