A Recommendation Algorithm Trading off Performance and Privacy Protection

doi:10.7652/xjtuxb202107013

您当前的位置：

首页 >

文章列表页 >

A Recommendation Algorithm Trading off Performance and Privacy Protection

更新时间：2025-07-09

- A Recommendation Algorithm Trading off Performance and Privacy Protection
- Vol. 55, Issue 7, Pages: 117-123(2021)
- 作者机构：
  
  1. 西安交通大学电子与信息学部,西安,710049
  2. 西安交通大学陕西省天地网技术重点实验室,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202107013
  CLC： TP391
- Online First：10 July 2021，
  
  Published：2021
- 稿件说明：
移动端阅览
A Recommendation Algorithm Trading off Performance and Privacy Protection[J]. 2021, 55(7): 117-123.
DOI：

A Recommendation Algorithm Trading off Performance and Privacy Protection[J]. 2021, 55(7): 117-123. DOI： 10.7652/xjtuxb202107013.

摘要

针对推荐系统中的隐私保护问题

提出一种隐私保护参数与推荐精度的均衡优化模型。以在线学习资源推荐系统为例建立矩阵分解模型

分别在数据输入和模型训练两个模块引入差分隐私噪声扰动

研究隐私保护参数ε与推荐精度的关系

并针对在线学习推荐系统数据的隐式反馈特点

提出基于资源热度负采样算法。基于西安交通大学网络学院数据集

利用百度飞桨深度学习平台

使用均方根误差作为推荐精度评价指标进行实验

结果表明:对数据基于资源热度负采样后

推荐精度更高; 推荐精度与ε的倒数成正比关系

输入扰动和模型扰动分别在均方根误差取值不高于2.0和1.3且ε取值7和3时均衡效果最优; 在相同ε下

当ε≤5时

模型扰动的差分隐私推荐算法推荐精度高于输入扰动的差分隐私推荐算法。

Abstract

Aiming at the problem of privacy protection in recommender systems

a balanced optimization model between privacy protection parameters and recommendation accuracy is proposed. Taking online learning resource recommender system as an example

this paper builds a matrix factorization model and studies the relationship between privacy protection parameters and recommendation accuracy after introducing differential privacy noise into data input module and model training module. According to the implicit feedback characteristics of online learning resource recommender system data

a negative sampling algorithm by resource-popularity is proposed. The experiment is based on the original and balanced data from the Network College of Xi'an Jiaotong University using Baidu PaddlePaddle platform

and the root mean square error is used as an evaluation index to measure the recommendation accuracy. The results show that negative sampling makes higher prediction accuracy than the original. And the recommendation prediction accuracy is directly proportional to the reciprocal of differential privacy protection parameter. When RMSEs are less than 2.0 and 1.3 and the privacy protection parameter is 7 and 3

respectively

both the input-based algorithm and the model-based algorithm achieve best balance. Moreover

when the privacy protection parameter is no more than 5

the model-based algorithm has a higher recommendation accuracy than the input-based algorithm.

关键词

Keywords

references

CALANDRINO J, KILZER A, NARAYANAN A, et al. You might also like: privacy risks of collaborative filtering [C]∥2011 IEEE Symposium on Security and Privacy. Piscataway, NJ, USA: IEEE, 2011: 231-246.

ABEL F, DELDJOO Y, KOHLSDORF D, et al. Recsyschallenge 2017: offline and online evaluation [C]∥Eleventh ACM Conference on Recommender Systems. New York, USA: ACM, 2017: 372-373.

KREN M, KOS A, ZHANG Y, et al. Public interest analysis based on implicit feedback of IPTV users [J]. IEEE Transactions on Industrial Informatics, 2017, 13(4): 2077-2086.

ZHANG J, CHONG W, LIU O, et al. Improving video recommendation systems from implicit feedback in the e-marketing environment [C]∥International Multi Conference of Engineers and Computer Scientists. Hong Kong, China: IAENG, 2017: 723-727.

周俊, 董晓蕾, 曹珍富. 推荐系统的隐私保护研究进展 [J]. 计算机研究与发展, 2019, 56(10): 2033-2048.

ZHOU Jun, DONG Xiaolei, CAO Zhenfu. Research progress of privacy protection in recommendation system [J]. Computer Research and Development, 2019, 56(10): 2033-2048.

谭作文, 张连福. 机器学习隐私保护研究综述 [J]. 软件学报, 2020, 31(7): 2127-2156.

TAN Zuowen, ZHANG Lianfu. A survey of privacy protection in machine learning [J]. Journal of Software, 2020, 31(7): 2127-2156.

熊平, 朱天清, 王晓峰. 差分隐私保护及其应用 [J]. 计算机学报, 2014, 37(1): 101-122.

XIONG Ping, ZHU Tianqing, WANG Xiaofeng. Differential privacy protection and its application [J]. Journal of Computer Science, 2014, 37(1): 101-122.

RENDLE S, KRICHENE W, ZHANG L, et al. Neural collaborative filtering vs matrix factorization revisited [C]∥Fourteenth ACM Conference on Recommender Systems. New York, USA: ACM, 2020: 240-248.

STEPHANE M, FALTINGS B, SCHICKEL V. Context-tree recommendation vs matrix-factorization: algorithm selection and live users evaluation [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2019(33): 9534-9540.

MENG Xuying, WANG Suhang, SHU Kai, et al. Personalized privacy preserving social recommendation [C]∥Proceedings of the 32nd AAAI Conference on Artificial Intelligence. New York, USA: ACM, 2018: 1-8.

FENG Pei, ZHU Haiping, LIU Yu, et al. Differential privacy protection recommendation algorithm based on student learning behavior [C]∥2018 IEEE 15th International Conference on e-Business Engineering(ICEBE). Piscataway, NJ, USA: IEEE, 2018: 285-288.

陆艺, 曹健. 面向隐式反馈的推荐系统研究现状与趋势 [J]. 计算机科学, 2016, 43(4): 7-15.

LU Yi, CAO Jian. Research status and trend of implicit feedback-oriented recommendation system [J]. Computer Science, 2016, 43(4): 7-15.

RICCI F, ROKACH L, SHAPIRA B, et al. Recommendation system: technology, evaluation and efficient algorithm [M]. Beijing, China: Machinery Industry Press, 2015.

KOREN Y, BELL R, VOLINSKY C. Matrix factorization techniques for recommender systems [J]. Computer, 2009, 42(8): 30-37.

SHOKRI R, STRONATI M, SONG C, et al. Membership inference attacks against machine learning models [J]. Cryptography and Security, 2017, 41(5): 3-18.

FREDRIKSON M, JHA S, RISTENPART T. Model inversion attacks that exploit confidence information and basic countermeasures [C]∥The 22nd ACM SIGSAC Conference. New York, USA: ACM, 2015: 1322-1333.

KAWALE J, BUI H, KVETON B, et al. Efficient Thompson sampling for online matrix-factorization recommendation [C]∥Neural Information Processing Systems(NIPS 2015). Cambridge, MA, USA: MIT Press, 2015: 1297-1305.

ZHU Haiping, LIU Yu, TIAN Feng, et al. A cross-curriculum video recommendation algorithm based on a video-associated knowledge map [J]. IEEE Access, 2018, 6(1): 57562-57571.

郭贵冰. 推荐系统进展: 方法与技术 [M]. 北京: 科学出版社, 2019: 37-38.

BERLIOZ A, FRIEDMAN A, KAAFAR M, et al. Applying differential privacy to matrix factorization [C]∥Proceedings of the 9th ACM Conference on Recommender Systems. New York, USA: ACM, 2015: 107-114.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Matrix Factorization Algorithm with Hybrid Implicit and Explicit Attributes for Recommender Systems

Singular Value Decomposition Recommendation Algorithm Considering User's Preference for Item Attributes

Deep Low-Rank Multi-View Subspace Clustering

A Fast Calculation Method for Thermal Flow Fields in Transformer Windings Based on PI-DeepONet

Effects of Steam Addition on the Swirl Combustion and Emission Characteristics of Cracked Ammonia

Related Author

赵长伟 1

彭勤科 1

张志勇 2

魏港明

刘真

李林峰

张猛

Related Institution

School of Electronic and Information Engineering, Xi'an Jiaotong University

School of Information and Engineering, Henan University of Science and Technology

School of Computer and Information Technology, Beijing Jiaotong University

School of Software Engineering, Xi'an Jiaotong University

⁰