A Hybrid Jiang Conrath Product Recommendation System for E-commerce in The Case of Data Sparsity
Abstract
Collaborative Filtering (CF) is commonly used in e-commerce recommendation systems, but it has limitations in cold-start conditions and data sparsity. This study implements the Hybrid Jiang-Conrath method to address these issues. This approach combines the semantic similarity of WordNet-based Jiang-Conrath Similarity (JCN) with the behavioral similarity of CF. JCN evaluates how conceptually similar the definitions of product categories were. Both approaches were combined using contribution values. The best contribution value across four datasets was α = 0.1 (10% CF, 90% JCN). The Hybrid model outperformed CF in the Industrial & Scientific dataset (MAE 0.64096, RMSE 0.88609). In contrast, the User-Based model achieved the lowest errors in Grocery & Gourmet (MAE 0.94661, RMSE 1.42904) and Video Games (MAE 0.04194, RMSE 0.04194). The Musical Instrument dataset showed comparable results between Item-Based (MAE 0.63247) and User-Based (RMSE = 1.05750) methods. Overall, the Hybrid model demonstrated better stability across diverse data sets. Compared to regular CF that offers only 31 products, hybrid Jiang-Conrath can generate more predictions for 65 product recommendations
References
Al-Bashiri, H., Abdulgabber, M. A., Romli, A., & Hujainah, F. (2017). Collaborative filtering recommender system: Overview and challenges. Advanced Science Letters, 23(9), 9045–9049. https://doi.org/10.1166/asl.2017.10020
Erlangga, E., & Sutrisno, H. (2020). Jurnal Manajemen Sistem Informasi dan Teknologi Sistem Rekomendasi Beauty Shop Berbasis Collaborative Filtering (Vol. 10, Issue 2).
Fahrurrozi, I., & Galuh Nurwana, A. (2019). PERBANDINGAN METODE COLLABORATIVE FILTERING DAN HYBRID JIANG-CONRATH (Vol. 17, Issue 1). http://wordnet.princeton.edu/
Hariri, F. R., & Rochim, L. W. (2022). Sistem Rekomendasi Produk Aplikasi Marketplace Berdasarkan Karakteristik Pembeli Menggunakan Metode User Based Collaborative Filtering. Teknika, 11(3), 208–217. https://doi.org/10.34148/teknika.v11i3.538
Hodson, T. O. (2022). Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not. In Geoscientific Model Development (Vol. 15, Issue 14, pp. 5481–5487). Copernicus GmbH. https://doi.org/10.5194/gmd-15-5481-2022
Mah, N., Ayyiyah, K., Kusumaningrum, R., & Rismiyati, R. (2023). FILM RECOMMENDER SYSTEM MENGGUNAKAN METODE NEURAL COLLABORATIVE FILTERING FILM RECOMMENDER SYSTEM USING NEURAL COLLABORATIVE FILTERING METHOD. 10(3), 699–708. https://doi.org/10.25126/jtiik.2023106616
Ramadhan Putra, K., & Fathur Rahman, I. (2024). MIND (Multimedia Artificial Intelligent Networking Database Pemanfaatan Metode Collaborative Filtering dengan Algoritma KNN pada Sistem Rekomendasi Produk. Journal MIND Journal | ISSN, 9(1), 113–123. https://doi.org/10.26760/mindjournal.v9i1.113-123
Ritdrix, A. H., & Wirawan, P. W. (2018). SISTEM REKOMENDASI BUKU MENGGUNAKAN METODE ITEM-BASED COLLABORATIVE FILTERING. In Sistem Rekomendasi Buku Menggunakan Metode… 24 Jurnal Masyarakat Informatika (Vol. 9, Issue 2).
Rosita, A., Puspitasari, N., & Kamila, V. Z. (2022). REKOMENDASI BUKU PERPUSTAKAAN KAMPUS DENGAN METODE ITEM-BASED COLLABORATIVE FILTERING. Sebatik, 26(1), 340–346. https://doi.org/10.46984/sebatik.v26i1.1551
Al-Hassan, M., Al-Khlifat, A., & Al-Ayyoub, M. (2024). An improved fusion-based semantic similarity measure for effective collaborative filtering recommendations. SN Computer Science, 5(2), 217. https://doi.org/10.1007/s42979-024-02417-4
Hou, Y., Li, J., He, Z., Yan, A., Chen, X., & McAuley, J. (2024). Bridging language and items for retrieval and recommendation (BLaIR). arXiv preprint arXiv:2403.03952. https://arxiv.org/abs/2403.03952
McAuley, J., & Li, J. (2023). Amazon Reviews 2023 Dataset. McAuley Lab, UCSD. https://huggingface.co/datasets/McAuley-Lab/Amazon-Reviews-2023
Movafegh, Z., Ghasemi, A., & Shiri, M. E. (2023). Improving collaborative recommender systems using hybrid model- and memory-based approaches. Engineering Applications of Artificial Intelligence, 123, 106285. https://doi.org/10.1016/j.engappai.2023.106285
Elahi, M., Bazarganigilani, M., & Cremonesi, P. (2023). A hybrid recommendation by incorporating the sentiment of user-generated content. Information Sciences, 642, 119152. https://doi.org/10.1016/j.ins.2023.119152
Drammeh, B., Kim, J., & Kim, Y. (2023). Enhancing neural collaborative filtering using hybrid deep feature extraction. PeerJ Computer Science, 9, e1505. https://doi.org/10.7717/peerj-cs.1505
Jeevamol, J., & Manjusha, K. (2021). An ontology-based hybrid e-learning content recommender system. Education and Information Technologies, 26, 7299–7322. https://doi.org/10.1007/s10639-021-10637-1
Ayemowa, M. O., Uduak, J., & Iorkyase, D. (2024). A systematic review of deep learning methods in recommender systems. Machine Learning with Applications, 17, 100535. https://doi.org/10.1016/j.mlwa.2024.100535
Li, D., Lian, J., Zhang, L., Ren, K., Lu, T., Wu, T., & Xie, X. (2024). Recommender Systems: Frontiers and Practices. Springer. https://doi.org/10.1007/978-981-97-0931-0
Bischof, B., & Yee, H. (2024). Building Recommendation Systems in Python and JAX. O’Reilly Media.
Zhang, X., Xu, Y., & Wang, T. (2022). Improving recommendation accuracy with hybrid semantic similarity and collaborative filtering. IEEE Access, 10, 56412–56425. https://doi.org/10.1109/ACCESS.2022.3179021
Yulianti, L., & Santoso, B. (2022). Sistem rekomendasi produk e-commerce menggunakan hybrid collaborative filtering. Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK), 9(5), 987–994. https://doi.org/10.25126/jtiik.20229654
Putra, D. R., & Saputra, A. (2021). Penerapan metode collaborative filtering pada sistem rekomendasi produk fashion berbasis web. Jurnal RESTI, 5(4), 698–706. https://doi.org/10.29207/resti.v5i4.3180
Ramadhani, S., & Wijaya, M. (2020). Sistem rekomendasi e-commerce menggunakan hybrid filtering untuk mengatasi sparsity data. Jurnal Teknologi dan Sistem Komputer, 8(2), 103–110. https://doi.org/10.14710/jtsiskom.8.2.2020
Fathurrahman, A., & Firmansyah, D. (2021). Evaluasi model rekomendasi menggunakan metrik MAE dan RMSE. Jurnal Teknologi dan Sistem Komputer, 9(3), 255–262. https://doi.org/10.14710/jtsiskom.9.3.2021
Barros, M., Oliveira, M., & Pereira, A. (2021). Hybrid semantic recommender system for compound discovery. Journal of Cheminformatics, 13(1), 47. https://doi.org/10.1186/s13321-021-00550-8
Media Indonesia. (2020, September). Article from Media Indonesia. https://www.mediaindonesia.com
AMZScout. (2025, October). Amazon statistics and market insights. https://www.amzscout.net
