Name
Abstract | ||
|---|---|---|
Online shopping is a common shopping style for human being nowadays. Rating mechanisms usually exist in most of the shopping sites. Therefore, predicting which products a customer is going to buy next from the rating information becomes possible, making recommender systems important for online shopping. The success of an online shopping site can be dominated by the quality of the recommender system involved. One factor for the quality is whether the user preference can be taken into account in the recommender system. A method that examines if a user's rating truly reflects a product's quality was proposed by Allahbakhsh and Lgnjatovic. By giving different weights to users' ratings according to the examined results, products are recommended to interested users automatically. However, the method does not consider that preferences are different among different individual users. In this paper, we propose a method which is also based on the examined results, but takes personal preference into account as well. As a result, different recommendations are preferably provided to different users. Experimental results are shown to demonstrate the effectiveness of our proposed method. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/ICInfA.2016.7831992 | 2016 IEEE INTERNATIONAL CONFERENCE ON INFORMATION AND AUTOMATION (ICIA) |
Keywords | Field | DocType |
machine learning, online rating, recommender system | Recommender system,World Wide Web,Computer science,Cluster analysis | Conference |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Chie-Hong Lee | 1 | 4 | 1.40 |
| Yann-Yean Su | 2 | 8 | 1.14 |
| Pang-Ming Chu | 3 | 0 | 1.01 |
| Shie-Jue Lee | 4 | 48 | 5.11 |