| With the rapid development of information technology.LBS provides more and more information at the same time.The recommendation system based on the location is the most effective way to solve this problem.However,in the process of the recommendation,the privacy of the user is likely to be threatened.In this paper,This paper mainly discusses the recommendation algorithms and related privacy issues in location based on the recommender systems:(1)Presenting a privacy preserving method for sensitive location.The traditional location based recommender system mostly adopts the collaborative filtering algorithm,which is used to calculate the similarity between user and location.But they ignore the influence of location on user privacy.To solve this problem,This paper presents a privacy preserving method for sensitive location,and prove that this method meets the requirements of privacy protection and does not affect the results(2)Presenting a privacy preserving method for independent location.In the process of the service,the similarity between users is calculated to provide recommendation results.But the systems tend to focus on the user similarity based on the similarity,but ignore the independent location,namely whether the location only belongs to a user,if such location data is recommended to other users,it will cause unnecessary loss of privacy.In order to solve this problem,this paper proposes a privacy detection method based on independent location,which protects the privacy of users.This paper also does the experimental to analyse the accuracy of the recommendation.(3)Optimizing the recommendation algorithm based on privacy preserving.In the process of the recommendation,according to the collaborative filtering algorithm based on the location,This paper designs different parameters,using the RMS error to indicate the accuracy of recommendation.Finally,This paper gets the best results of the number of iterations,the feature vectors and so on. |
PDF Full Text Request