Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2025
|
| Online Access: | https://doi.org/10.5281/zenodo.15129756 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>—The growing need for accurate recommendations<br> has motivated the devel opment of hybrid recommendation<br> systems. In this project, we propose a sys tem that integrates<br> collaborative filtering, content-based filtering, and knowledge<br> graph approaches to enhance recommendation accuracy and<br> address limitations such as cold-start and data sparsity. The<br> system uses a multi-step method: col laborative filtering predicts<br> user preferences based on similar users, content-based filtering<br> analyses item attributes, and a knowledge graph refines recom<br>mendations by leveraging relational data. The hybrid approach<br> offers advantages like more personalized recommendations and<br> better handling of sparse data. However, it has the drawback of<br> increased computational complexity due to integrating multiple<br> algorithms. In conclusion, the system aims to effectively combines<br> the strengths of different recommendation techniques, to provide<br> more precise and relevant suggestions to users.</p>