DocumentCode :
2400114
Title :
Anomaly Detection in Feedback-based Reputation Systems through Temporal and Correlation Analysis
Author :
Yuhong Liu ; Yan Sun
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Rhode Island, Kingston, RI, USA
fYear :
2010
fDate :
20-22 Aug. 2010
Firstpage :
65
Lastpage :
72
Abstract :
As the value of reputation systems is widely recognized, the incentive to manipulate such systems is rapidly growing. We propose TAUCA, a scheme that identifies malicious users and recovers reputation scores from a novel angle: combination of temporal analysis and user correlation analysis. Benefiting from the rich information in the time-domain, TAUCA identifies the products under attack, the time when attacks occur, and malicious users who insert dishonest ratings. TAUCA and two other representative schemes are tested against real user attack data collected through a cyber competition. TAUCA demonstrates significant advantages. It largely improves the detection rate and reduces the false alarm rate in the detection of malicious users. It also effectively reduces the bias in the recovered reputation scores.
Keywords :
Internet; security of data; Internet; TAUCA; anomaly detection; feedback-based reputation systems; malicious user identification; reputation score recovery; temporal analysis; user correlation analysis; Boosting; Correlation; Delay; Detectors; Indexes; Silicon; Time domain analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Social Computing (SocialCom), 2010 IEEE Second International Conference on
Conference_Location :
Minneapolis, MN
Print_ISBN :
978-1-4244-8439-3
Type :
conf
DOI :
10.1109/SocialCom.2010.19
Filename :
5590838
Link To Document :
بازگشت