An extended trust and distrust network-based dual fuzzy recommendation model and its application based on user-generated content
Sichao Chen,Shengjia Zhou
DOI: https://doi.org/10.1016/j.eswa.2024.123360
IF: 8.5
2024-02-07
Expert Systems with Applications
Abstract:The social network-based recommendation algorithm can not only effectively solve the information overload problem, but also fully exploit the personalized preferences of users. However, current studies fail to consider the distrust feature and propagation characteristic of social networks. In addition, methods for quantifying user preferences and social relationships are deficient and inconsistent with reality. Therefore, an extended trust and distrust network-based dual fuzzy recommendation model (ETD-DFR) is proposed. First, the trust and distrust network is constructed, and then the network is extended based on the social balance theory; secondly, the weighted PageRank algorithm is utilized to calculate the trust and distrust between users; Further, a dual fuzzy framework is proposed to represent the fuzzy features of user preferences as well as trust and distrust; In addition, Louvain's algorithm is employed to partition the network into subgroups to quickly find trusted users for recommendation. An empirical analysis is conducted based on the dataset of consumer review websites, and the experimental results show that the performance of the proposed ETD-DFR model is excellent. Finally, comparative and sensitivity analyses are constructed to further verify the proposed ETD-DFR model is efficient and robust.
computer science, artificial intelligence,engineering, electrical & electronic,operations research & management science