Abstract | ||
---|---|---|
For years, recommender systems (RS) have emerged as a powerful tool to enable users to find appropriate information according to their needs. Different recommendation methods have been proposed and can be categorized as collaborative filter, content-based, and Hybrid/Ensemble approach. However, the exponential growth of digital information in the recent decades often referred to Big Data, poses new challenges for the current RS. Following this spirit, our work proposes a novel fast clustering-based Recommendation method (denoted as FCR) designed on top of Apache Spark. Comprehensive experiments on a real-world dataset have verified the advantages of our proposed method. It is effective in alleviating the problem of data sparsity and item cold-start. The training and inference time is quick while the slight increase of Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) is acceptable. |
Year | DOI | Venue |
---|---|---|
2022 | 10.23919/ICACT53585.2022.9728770 | 2022 24TH INTERNATIONAL CONFERENCE ON ADVANCED COMMUNICATION TECHNOLOGY (ICACT): ARITIFLCIAL INTELLIGENCE TECHNOLOGIES TOWARD CYBERSECURITY |
Keywords | DocType | ISSN |
Recommender System, Big Data, Clustering-based Recommendation, Item cold-start, Data sparsity, Apache Spark | Conference | 1738-9445 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hong-Quan Do | 1 | 0 | 1.69 |
T. H-An Nguyen | 2 | 0 | 0.34 |
Quoc-Anh Nguyen | 3 | 0 | 0.34 |
Trung-Hieu Nguyen | 4 | 0 | 0.34 |
Viet-Vu Vu | 5 | 0 | 1.69 |
Cuong Le | 6 | 0 | 0.68 |