TY - JOUR
T1 - Latest trends of security and privacy in recommender systems
T2 - A comprehensive review and future perspectives
AU - Himeur, Yassine
AU - Sohail, Shahab Saquib
AU - Bensaali, Faycal
AU - Amira, Abbes
AU - Alazab, Mamoun
N1 - Funding Information:
This paper was made possible by National Priorities Research Program (NPRP ) grant no. 10-0130-170288 from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors.
Publisher Copyright:
© 2022
PY - 2022/7
Y1 - 2022/7
N2 - With the widespread use of Internet of things (IoT), mobile phones, connected devices and artificial intelligence (AI), recommender systems (RSs) have become a booming technology because of their capability to analyze big data and shape users’ habits through well-designed, contextual, and engaging recommendations. Novel generations of RSs have been developed based on the latest AI and machine learning (ML) technologies such as big data RSs, ML-based RSs, explainable RSs, fusion-based RSs, etc. However, the characteristics of modern RSs raise new security and privacy issues because of the sensitivity of users’ data and its vulnerability to being illegally accessed. Moreover, there is a lack of thorough reviews that explain the current privacy and security challenges in RSs and where the actual research is heading. To overcome these issues, this paper sheds light on the existing security and privacy concerns in modern RSs. It provides a comprehensive survey of recent research efforts on security and privacy preservation in RSs. Typically, the security and privacy aspects in advanced RSs and the latest contributions are first discussed based on a well-defined taxonomy. Next, the applications of secure and privacy-preserving RSs are studied. Moving forward, a critical analysis is conducted to (i) highlight the merits and drawbacks of existing frameworks and (ii) draw the essential findings. Lastly, future directions that attract significant research and development attention are explained.
AB - With the widespread use of Internet of things (IoT), mobile phones, connected devices and artificial intelligence (AI), recommender systems (RSs) have become a booming technology because of their capability to analyze big data and shape users’ habits through well-designed, contextual, and engaging recommendations. Novel generations of RSs have been developed based on the latest AI and machine learning (ML) technologies such as big data RSs, ML-based RSs, explainable RSs, fusion-based RSs, etc. However, the characteristics of modern RSs raise new security and privacy issues because of the sensitivity of users’ data and its vulnerability to being illegally accessed. Moreover, there is a lack of thorough reviews that explain the current privacy and security challenges in RSs and where the actual research is heading. To overcome these issues, this paper sheds light on the existing security and privacy concerns in modern RSs. It provides a comprehensive survey of recent research efforts on security and privacy preservation in RSs. Typically, the security and privacy aspects in advanced RSs and the latest contributions are first discussed based on a well-defined taxonomy. Next, the applications of secure and privacy-preserving RSs are studied. Moving forward, a critical analysis is conducted to (i) highlight the merits and drawbacks of existing frameworks and (ii) draw the essential findings. Lastly, future directions that attract significant research and development attention are explained.
KW - Authentication
KW - Blockchain
KW - Malicious attacks
KW - Recommender systems
KW - Security and privacy
KW - Trust
UR - http://www.scopus.com/inward/record.url?scp=85130092496&partnerID=8YFLogxK
U2 - 10.1016/j.cose.2022.102746
DO - 10.1016/j.cose.2022.102746
M3 - Review article
AN - SCOPUS:85130092496
SN - 0167-4048
VL - 118
SP - 1
EP - 28
JO - Computers and Security
JF - Computers and Security
M1 - 102746
ER -