| Autor: |
Vihanga, W. G. Dulakshi, Barraza-Urbina, Andrea, Jayaratne, K. L. |
| Předmět: |
|
| Zdroj: |
Annual International Conference on Information Technology & Applications; 2016, p53-58, 6p |
| Abstrakt: |
Recommender Systems (RS) are progressively gaining more prominence due to the escalation of information accessible through the web. The main quality aspect of RS has long been regarded as the accuracy of recommendations, which is associated with how relevant the recommendations are to each individual user. However, diversity is receiving heightened attention as an important dimension of the quality of recommendations. Although diversity is a desired aspect, the preference for diversity is not the same for each individual. Since personalization of diversification has been overlooked, we propose a technique to carry-out the diversification process in a personalized manner while allowing controlling the trade-off between diversity and accuracy. Moreover, we also introduce a novel recommendation technique which uses the past preferences of a user and the ratings of experienced item category experts in recommendation generation process. We integrate these two techniques to provide a more personalized diversification experience. Furthermore, unlike other techniques, our approach can promote both novel and relevant items and also make the diversification process personalized. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
