Hybridizing Social Filtering for Recommender Systems

Publication Type:
Conference Proceeding
Citation:
Advances in Intelligent Systems and Computing, 2014, 277 pp. 273 - 285
Issue Date:
2014-01-01
Metrics:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnailpaper.pdfAccepted Manuscript version494.11 kB
Adobe PDF
Users send requests to recommender systems for getting suggested products or services. Collaborative filtering is a popular technique for making such suggestions efficiently, but it suffers from a drawback known as "cold-start" problem. Social filtering may succeed for such users, since it utilize the extra social relations of users. It gives us opportunities to eliminate the limitations by hybridizing social filtering into traditional collaborative filtering. To handle this issue, differing from previous fusion models that only combine the final results, this paper proposed a new neighborhood fusion model to make hybridization at an earlier and deeper stage. Experiment-based comparative analyses are also conducted. The results show that our model is of a higher recommendation quality, on different datasets. © Springer-Verlag Berlin Heidelberg 2014.
Please use this identifier to cite or link to this item: