Development Of A Collaborative Recommendation System Based on Singular Value Decomposition (SVD) on E-Commerce Data
DOI:
https://doi.org/10.30871/jaic.v9i6.11688Keywords:
Collaborative Filtering, E-commerce, Matrix Factorization, Recommendation Systems, Singuler Value DecompositionAbstract
Recommendation systems (RS) are vital tools for mitigating information overload and data sparsity challenges in modern e-commerce platforms. This study focuses on developing and evaluating a Collaborative Filtering (CF) model utilizing Singular Value Decomposition (SVD) as a Matrix Factorization technique, applied to the publicly available E-commerce dataset. The dataset, comprising nine interconnected transactional tables, presents significant data sparsity due to limited explicit user ratings relative to the vast product catalog. The SVD model was implemented to decompose the highly sparse User-Item interaction matrix into lower-rank latent factor matrices, thereby capturing underlying purchasing patterns and user preferences. The model's performance was rigorously validated using k-fold cross-validation and assessed via standard accuracy metrics: Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The experimental results demonstrated high predictive accuracy, achieving an RMSE of 1.25 and an MAE of 0.98. These findings indicate that the SVD model effectively overcomes the sparsity challenge inherent in large-scale e-commerce transactional data, providing robust prediction capabilities that surpass established industry benchmarks (e.g., RMSE » 1.31, MAE » 1.04 found in similar studies). The successful implementation validates SVD as a highly effective approach for generating personalized, high-quality product recommendations, offering substantial business implications for enhancing customer engagement and maximizing Average Order Value (AOV)
Downloads
References
[1] X. Luo, M. Zhou, Y. Xia, and Q. Zhu, “An efficient non-negative matrix-factorization-based approach to collaborative filtering for recommender systems,” IEEE Trans Industr Inform, vol. 10, no. 2, pp. 1273–1284, 2014, doi: 10.1109/TII.2014.2308433.
[2] D. Cai, X. He, J. Han, and T. S. Huang, “Graph regularized nonnegative matrix factorization for data representation,” IEEE Trans Pattern Anal Mach Intell, vol. 33, no. 8, pp. 1548–1560, 2011, doi: 10.1109/TPAMI.2010.231.
[3] E. Frolov and I. Oseledets, “Tensor methods and recommender systems,” Wiley Interdiscip Rev Data Min Knowl Discov, vol. 7, no. 3, May 2017, doi: 10.1002/WIDM.1201.
[4] V. N. Ioannidis, A. S. Zamzam, G. B. Giannakis, and N. D. Sidiropoulos, “Coupled Graphs and Tensor Factorization for Recommender Systems and Community Detection,” IEEE Trans Knowl Data Eng, vol. 33, no. 3, pp. 909–920, Mar. 2021, doi: 10.1109/TKDE.2019.2941716.
[5] S. Shlien, “A Method for Computing the Partial Singular Value Decomposition,” IEEE Trans Pattern Anal Mach Intell, vol. PAMI-4, no. 6, pp. 671–676, 1982, doi: 10.1109/TPAMI.1982.4767324.
[6] M. Gong, X. Jiang, H. Li, and K. C. Tan, “Multiobjective Sparse Non-Negative Matrix Factorization,” IEEE Trans Cybern, vol. 49, no. 8, pp. 2941–2954, Aug. 2019, doi: 10.1109/TCYB.2018.2834898.
[7] J. Vinagre, A. M. Jorge, C. Rocha, and J. Gama, “Statistically Robust Evaluation of Stream-Based Recommender Systems,” IEEE Trans Knowl Data Eng, vol. 33, no. 7, pp. 2971–2982, Jul. 2021, doi: 10.1109/TKDE.2019.2960216.
[8] I. Ramírez, “Binary Matrix Factorization via Dictionary Learning,” IEEE Journal on Selected Topics in Signal Processing, vol. 12, no. 6, pp. 1253–1262, Dec. 2018, doi: 10.1109/JSTSP.2018.2875674.
[9] H. Shin, S. Kim, J. Shin, and X. Xiao, “Privacy Enhanced Matrix Factorization for Recommendation with Local Differential Privacy,” IEEE Trans Knowl Data Eng, vol. 30, no. 9, pp. 1770–1782, Sep. 2018, doi: 10.1109/TKDE.2018.2805356.
[10] T. T. Wong and P. Y. Yeh, “Reliable Accuracy Estimates from k-Fold Cross Validation,” IEEE Trans Knowl Data Eng, vol. 32, no. 8, pp. 1586–1594, Aug. 2020, doi: 10.1109/TKDE.2019.2912815.
[11] T. T. Wong and N. Y. Yang, “Dependency Analysis of Accuracy Estimates in k-Fold Cross Validation,” IEEE Trans Knowl Data Eng, vol. 29, no. 11, pp. 2417–2427, Nov. 2017, doi: 10.1109/TKDE.2017.2740926.
[12] T. Subaranjani, S. Kannaiyan, S. Parvathy and P. R. Shwetha, "Enhancing E-commerce Fashion Sales through Personalized Recommendation Systems," 2024 11th International Conference on Computing for Sustainable Global Development (INDIACom), New Delhi, India, 2024, pp. 290-296, doi: 10.23919/INDIACom61295.2024.10498171.
[13] R. Pari, K. Pallavi and P. Amareshwaran, "An Effective Collaborative Filtering Technique Using Single Value Decomposition for Book Recommendation," 2025 IEEE International Conference on Contemporary Computing and Communications (InC4), Bangalore, India, 2025, pp. 1-6, doi: 10.1109/InC465408.2025.11256220.
[14] M. Pylypchuk, N. Porplytsya, I. Stasiv, L. Honchar, V. Sopiha and I. Bondarenko, "Optimisation of SVD++ Method based on Adam’s Algorithm for Small E-Commerce Platforms," 2024 14th International Conference on Advanced Computer Information Technologies (ACIT), Ceske Budejovice, Czech Republic, 2024, pp. 318-321, doi: 10.1109/ACIT62333.2024.10712569.
[15] F. Islam, M. S. Arman, N. Jahan, M. H. Sammak, N. Tasnim and I. Mahmud, "Model and Popularity Based Recommendation System- A Collaborative Filtering Approach," 2022 13th International Conference on Computing Communication and Networking Technologies (ICCCNT), Kharagpur, India, 2022, pp. 1-5, doi: 10.1109/ICCCNT54827.2022.9984348.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Galih Mahalisa, Silvia Ratna, M. Muflih
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
