Memory-based collaborative filtering (i.e., MCF) is regarded as an effective technique to recommend appropriate services to target users. However, if recommendation data are very sparse in the edge environment, traditional MCF-based recommendation methods probably cannot output any recommended item (or service), i.e., a cold-start recommendation problem occurs. To cope with this cold-start problem, we propose an intelligent recommendation method named Inverse_CF_Rec. Concretely, for a target user, we first search for his/her opposite users (together referred to as "enemy" hereafter); afterward, we infer the possible friends of the target user indirectly according to Social Balance Theory; finally, optimal services are recommended to the target user based on the derived possible friends of the target user. The experiments are conducted on a real-world dataset WS-DREAM to validate the effectiveness and efficiency of our proposal. The experiment results show the advantages of our recommendation method in terms of recommendation accuracy and efficiency.
Bibliographical noteVersion archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.
- Service recommendation
- inverse collaborative filtering
- social balance theory
- Inverse collaborative filtering
- Social balance theory