Open AccessUser Abnormal Viewing Behavior Detection Method Based On Canopy-FCM and Improved Isolation Forest Algorithm
Author(s) -
Yu Chieh Wu,
Huiyong Liu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1792/1/012057
Subject(s) - computer science , anomaly detection , cluster analysis , iptv , algorithm , precision and recall , artificial intelligence , data mining , multimedia
In order to analyze the users’ viewing behavior more accurately, allow IPTV service providers to provide better services to users, and improve the user’s viewing experience, this paper proposes a detection method of abnormal user viewing behavior based on Canopy-FCM and improved isolation forest algorithm. First, the collected viewership data is preprocessed and stored with the user viewing behavior matrix, then the Canopy-FCM clustering algorithm is used to classify users with the same rating behavior. Then, an improved isolated forest algorithm is used to perform anomaly detection and analysis on the users classified above, and finally screen out abnormal user viewing behaviors. Finally, NMI indicators and ROC curves are used to evaluate the performance of the proposed algorithm to verify the feasibility of the proposed anomaly detection method. The results show that the method proposed in this paper has a high precision and recall rate for detecting users with abnormal ratings.