Abstract | ||
---|---|---|
This paper describes the design and implementation of a novel approach to dynamically adjust the recommendation list size for multiple preferences of a user. By considering users' earlier preferences, machine learning techniques are employed to estimate the optimal recommendation list size according to current conditions of users. The proposed approach has been evaluated on real-life data from grocery shopping domain by conducting a series of experiments. The results show that the proposed approach achieves better overall recommendation quality than the standard approach and it outperforms the benchmark method in efficiency by shortening the recommendation list while maintaining the effectiveness. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1007/978-3-319-44748-3_30 | ARTIFICIAL INTELLIGENCE: METHODOLOGY, SYSTEMS, AND APPLICATIONS, AIMSA 2016 |
Keywords | Field | DocType |
Top-N recommender systems, Recommendation list size, Recommendation length, Recommendation quality, Recommendation efficiency | Computer science,Grocery shopping,Artificial intelligence,Machine learning | Conference |
Volume | ISSN | Citations |
9883 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 10 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Peker, S. | 1 | 2 | 1.40 |
Altan Koçyigit | 2 | 22 | 8.09 |