A Collaborative Filtering Group Recommendation Method Based on Random Perturbation Technology

Abstract

The invention discloses a collaborative filtering group recommendation method based on random perturbation technology, and mainly aims at the problems that users may expose private information during the collaborative filtering group recommendation process, and proposes a privacy protection group recommendation method, which mainly includes proposing user Raw score data perturbation processing, user interest preference value, preference weight and frequency weight, process the user's original data through random disturbance technology to protect the user's privacy data, user interest preference value forms a common preference group, preference weight and frequency weight The degree weight is used to obtain the group preference value, and the item-based collaborative filtering is used to obtain the prediction score of the common preference group, which meets the needs of the group to obtain recommendations and protect the privacy data of users in the group.

Description

technical field [0001] The invention relates to the technical fields of privacy protection and recommendation systems, in particular to a collaborative filtering group recommendation method based on random perturbation technology. Background technique [0002] The emergence and development of Internet technology has made it possible for massive information in the network. In order to facilitate users to select the required information, people put forward the concept of recommendation system. The recommendation system is established by the recommendation service provider and realizes user preference prediction by collecting user data. The collected data may contain various private information, such as mailing address or order history, so the privacy of user preference data becomes a concern. With the development and gradual increase of network users, it is easier for users to form a group by sharing common hobbies and participate in related activities together, which stimulat...

AI Technical Summary

Problems solved by technology

[0003] Random perturbation is a method to protect privacy by adding random noise to the original data. The perturbed data generated by this method can be regarded as approximating the original data. The principle is to add perturbation noise that obeys a specific distribution to the original data. Although the fuzzy processing has changed the original data, it does not have a great impact on the results of some calculations, and the error can be widely used within a certain range.

Images