Name
Abstract | ||
|---|---|---|
The recommendation system as an effective tool is used to alleviate the information overload problem, and is being applied to personalised services. In recommendation, user's ratings as explicit feedback data can clearly express user's preference, however explicit feedback data has a natural defect that user's interest for an item would varies in context such as emotions, time etc. and the ratings could not reflect the changing. In this paper, a novel graph recommendation algorithm is presented based on user's trust relation that is regarded as implicit feedback data to calculate similarity to enhance the performance for Top-K recommendations. By evaluating the presented algorithm and compared to four competitive algorithms on the four real world datasets, the results show that the presented algorithm performs better than other algorithms in precision, recall and converge. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1504/IJIPT.2021.113906 | INTERNATIONAL JOURNAL OF INTERNET PROTOCOL TECHNOLOGY |
Keywords | DocType | Volume |
recommendation, graph-based algorithm, implicit feedback, social networks | Journal | 14 |
Issue | ISSN | Citations |
1 | 1743-8209 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Liejun Wang | 1 | 2 | 1.72 |
| Long Pan | 2 | 0 | 0.34 |
| Jiwei Qin | 3 | 0 | 2.70 |