A Hybrid Recommender System Using KNN and Clustering

Hao Fan, Kaijun Wu, Hamid Parvin, Akram Beigi () and Kim-Hung Pho
Additional contact information
Hao Fan: College of Information Technology, Shanghai Ocean University, Shanghai, P. R. China
Kaijun Wu: College of Information Technology, Shanghai Ocean University, Shanghai, P. R. China
Hamid Parvin: #x2020;Institute of Research and Development, Duy Tan University, Da Nang 550000, Vietnam‡Faculty of Information Technology, Duy Tan University, Da Nang 550000, Vietnam§Department of Computer Engineering, Nourabad Mamasani Branch, Islamic Azad University, Nourabad Mamasani, Iran
Akram Beigi: #xB6;Shahid Rajaee Teacher Training University, Tehran, Iran
Kim-Hung Pho: #x2225;Fractional Calculus, Optimization and Algebra Research Group, Faculty of Mathematics and Statistics, Ton Duc Thang University, Ho Chi Minh City, Vietnam

International Journal of Information Technology & Decision Making (IJITDM), 2021, vol. 20, issue 02, 553-596

Abstract: Recommender Systems (RSs) are known in the E-Commerce (EC) field. They are expected to suggest the accurate goods/musics/films/items to the consumers/clients/people/users. Recent Hybrid RSs (HRSs) have made us able to deal with the most important shortages of traditional Content-based F iltering (ConF) and Collaborative Filtering (ColF). Cold start, scalability and sparsity are the most important challenges to EC recommender systems (ECRS). HRSs combine ConF and ColF. While the RSs that are based on memory have high accuracy, they are not scalable. Contrarily, the RSs on the basis of models have low accuracy but high scalability. Thus, aiming at dealing with cold start, scalability and sparsity challenges, HRS is proposed to use both methods and also it has been evaluated on a real benchmark. An ontology, which is automatically created by an intelligently collected wordnet, has been employed in ConF segment of the proposed HRS. It has been automatically created and enhanced by an additional process. The functionality of the recommended framework has been superior to the performance of the state-of-the-art methods and the traditional ConF and ColF embedded in our method. Using a real dataset as a benchmark, the experimentations indicate that the proposed method not only has better performance but also has more efficacy rather than the state-of-the-art methods.

Keywords: Recommender system; wordnet; ontology; clustering; KNN (search for similar items in EconPapers)
Date: 2021
References: Add references at CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.worldscientific.com/doi/abs/10.1142/S021962202150005X
Access to full text is restricted to subscribers

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wsi:ijitdm:v:20:y:2021:i:02:n:s021962202150005x

Ordering information: This journal article can be ordered from

DOI: 10.1142/S021962202150005X

Access Statistics for this article

International Journal of Information Technology & Decision Making (IJITDM) is currently edited by Yong Shi

More articles in International Journal of Information Technology & Decision Making (IJITDM) from World Scientific Publishing Co. Pte. Ltd.
Bibliographic data for series maintained by Tai Tone Lim ().