Geographical information in a multi-domain recommender system
Document Type
Conference Proceeding
Publication Date
1-1-2014
Abstract
Multi-domain recommendation is more challenging than that in the traditional single-domain one. In our previous study on the cross-domain recommendation, we have uncovered the tradeoff between the accuracy and the coverage of the recommendation. Later, we have also reported our findings on uncovering the association between user’s interests of items across domains that are related to each other to a certain degree using another dataset collected from users with different demographic information. In this paper, we further discuss the comparison between our previous two experimental results and some practical implications in the design of recommendation systems.
Publication Title
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
First Page Number
315
Last Page Number
321
DOI
10.1007/978-3-319-11538-2_29
Recommended Citation
Tang, Tiffany Y.; Winoto, Pinata; and Ye, Robert Ziqin, "Geographical information in a multi-domain recommender system" (2014). Kean Publications. 1997.
https://digitalcommons.kean.edu/keanpublications/1997