IJBTT

Enhancing Movie Recommendation using Ensemble based Machine Learning Approach

Enhancing Movie Recommendation using Ensemble based Machine Learning Approach
	© 2024 by IJETT Journal
	Volume-72 Issue-10
	Year of Publication : 2024
	Author : Sneha Bohra, Amit Gaikwad
	DOI : 10.14445/22315381/IJETT-V72I10P119

How to Cite?
Sneha Bohra, Amit Gaikwad, "Enhancing Movie Recommendation using Ensemble based Machine Learning Approach," International Journal of Engineering Trends and Technology, vol. 72, no. 10, pp. 191-203, 2024. Crossref, https://doi.org/10.14445/22315381/IJETT-V72I10P119

Abstract
Traditional Recommendation Systems suffer from Concept Drift- a phenomenon that assumes that user preferences are static over time. To address this issue, there is a need for a recommendation algorithm that takes into account time-sensitive shifts in user preferences and offers relevant recommendations. This research work proposes an Ensemble-based Hybrid Recommendation System which incorporates temporal variations in user interests. The proposed system combines distinct algorithms such as Popularity, Clustering, Collaborative Matrix Factorization and Singular Value Decomposition (SVD). The movie recommendations obtained from these individual models are then combined and classified using Artificial Neural Network (ANN). User feedback on the presented batch of recommendations is then recorded, which contributes to the calculation of the relevancy factor for each batch of recommendations. Finally, the user is provided with relevant movie recommendations. In case of a lower relevancy factor, the recommendations are reclassified. The objective of the proposed system is to provide diverse recommendations to the user depending on his time-sensitive preferences. The novelty of the proposed research is the integration of the popular recommendation strategies and the incorporation of the user feedback mechanism on the presented recommendations. The proposed system is implemented on the standard movie dataset Movielens-25m and is evaluated using statistical performance metrics like RMSE and MAE. The experiments show that the integration of an Artificial Neural Network as a classifier to the Ensemble- Hybrid Recommendation Model demonstrates promising results in terms of providing relevant recommendations with 0.56 and 0.43 as RMSE and MAE values. The work illustrates the improvement in recommendation quality and increased adaptability to varying user preferences.

Keywords
Movie recommendation system, Collaborative filtering, Content-based filtering, Hybrid approach, Singular Value Decomposition (SVD), Neural Network (NN).

References
[1] R.N. Ravi Kumar, Sanjay Jain, and Manash Sarkar, “Efficient Hybrid Movie Recommendation System Framework Based on a Sequential Model,” International Journal of Intelligent Systems and Applications in Engineering, vol. 11, no. 9, pp. 145-155, 2023.
[Google Scholar] [Publisher Link]
[2] Chia-Chi Wu, Yen-Liang Chen, and Yi-Hsin Yeh, “A Deep Recommendation Model Considering the Impact of Time and Individual Diversity,” IEEE Transactions on Computational Social Systems, vol. 11, no. 2, pp. 2558-2569, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Hanrui Wu et al., “Cold-Start Next-Item Recommendation by User-Item Matching and Auto-Encoders,” IEEE Transactions on Services Computing, vol. 16, no. 4, pp. 2477-2489, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Luong Vuong Nguyen et al., “Extending Collaborative Filtering Recommendation Using Word Embedding: A Hybrid Approach,” Concurrency and Computation: Practice and Experience, vol. 35, no. 16, pp. 6232-6243, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Syjung Hwang, and Eunil Park, “Movie Recommendation Systems Using Actor-Based Matrix Computations in South Korea,” IEEE Transactions on Computational Social Systems, vol. 9, no. 5, pp. 1387-1393, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Syjung Hwang, Hyeongjin Ahn, and Eunil Park, “iMovieRec: A Hybrid Movie Recommendation Method Based on a User-Image-Item Model,” International Journal of Machine Learning and Cybernetics, vol. 14, no. 9, pp. 3205-3216, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Ayush Tanwar, and Dinesh Kumar Vishwakarma, “A Deep Neural Network-Based Hybrid Recommender System with User-User Networks,” Multimedia Tools and Application, vol. 82, no. 10, pp. 15613-15633, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Payam Bahrani et al., “A Hybrid Semantic Recommender System Enriched with an Imputation Method,” Multimedia Tools and Applications, vol. 83, pp. 15985-16018, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Zhu Wang et al., “VRConvMF: Visual Recurrent Convolutional Matrix Factorization for Movie Recommendation,” IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 6, no. 3, pp. 519-529, 2022.
[CrossRef] [Google Scholar] [Publisher Link]