This paper introduces PerDif; a novel framework for learning personalized difusions over item-to-item graphs for top-n recommendation. PerDif learns the teleportation probabilities of a time-inhomogeneous random walk with restarts capturing a user-specifc underlying item exploration process. Such an approach can lead to signifcant improvements in recommendation accuracy, while also providing useful information about the users in the system. Per-user ftting can be performed in parallel and very efciently even in large-scale settings. A comprehensive set of experiments on real-world datasets demonstrate the scalability as well as the qualitative merits of the proposed framework. PerDif achieves high recommendation accuracy, outperforming state-of-the-art competing approaches-including several recently proposed methods relying on deep neural networks.
|Original language||English (US)|
|Title of host publication||RecSys 2019 - 13th ACM Conference on Recommender Systems|
|Publisher||Association for Computing Machinery, Inc|
|Number of pages||9|
|State||Published - Sep 10 2019|
|Event||13th ACM Conference on Recommender Systems, RecSys 2019 - Copenhagen, Denmark|
Duration: Sep 16 2019 → Sep 20 2019
|Name||RecSys 2019 - 13th ACM Conference on Recommender Systems|
|Conference||13th ACM Conference on Recommender Systems, RecSys 2019|
|Period||9/16/19 → 9/20/19|
Bibliographical noteFunding Information:
This work was supported in part by NSF (1447788, 1704074, 1757916, 1834251), Army Research Ofce (W911NF1810344), Intel Corp, and the Digital Technology Center at the University of Minnesota. Access to research and computing facilities was provided by the Digital Technology Center and the Minnesota Supercomputing Institute.
- Item Models
- Random Walks
- Top-N Recommendation