A Movie Recommendation using Common Genre Relation on User-Item Subgroup
G. Suganeshwari1, S.P Syed Ibrahim2
1G. Suganeshwari, Department of Computing Science and Engineering, Vellore Institute of Technology, Chennai (Tamil Nadu), India.
2S. P. Syed Ibrahim, Department of Computing Science and Engineering, Vellore Institute of Technology, Chennai (Tamil Nadu), India.
Manuscript received on 14 December 2019 | Revised Manuscript received on 22 December 2019 | Manuscript Published on 31 December 2019 | PP: 193-199 | Volume-9 Issue-1S3 December 2019 | Retrieval Number: A10381291S319/19©BEIESP | DOI: 10.35940/ijeat.A1038.1291S319
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Movie recommendation system has played a vital role in retrieving the movies that are of interest to the user. Most of the traditional methods provide a unified recommendation without considering the individual preference of the user. To address this challenge, various recommender methods are currently employing side information like location, time, gender, and genre to provide a personalized recommendation. In this paper, we propose —Common Genre Relations (COGS), which incorporates the information on genre relationships between the movies. Meanwhile, the method reduces the search space for each user and helps to mitigate the sparsity problem. To improve the scalability, the methods are executed on user-item subgroups. Extensive experiments are conducted on a real-world dataset. The empirical analysis shows that the proposed method based on the graph model excels the accuracy at top-k than the state-of-art collaborative filtering methods.
Keywords: Collaborative Filtering, Data Sparsity, Genre Relation, Movie Recommendation System, Random Walk.
Scope of the Article: Data Analytics