Improved Next Item Recommendation for Long Tail Products with Machine Learning

Abstract

Many different types of products can be sold on electronic commerce platforms. Products can be sold regardless of where customers are. The recommendation system on these platforms plays a critical role in selecting and displaying interesting products for users. In the study, the products to be purchased next to the customers were recommended in the most accurate way. For this, machine learning algorithms were used and the results were compared. The singular value decomposition (SVD) method has achieved more successful results.

Keywords

• Long tail products
• singular value decomposition
• k-nearest neighbor
• matrix factorization
• next item recommendation

Makine Öğrenimi ile Uzun Kuyruk Ürünler için İyileştirilmiş Sonraki Öğe Önerisi

Abstract

Elektronik ticaret platformlarında birçok farklı ürün türü müşterilerin nerede olduklarından bağımsız olarak satılabilmektedir. Bu platformlarda bulunan öneri sistemi kullanıcılar için ilgi çekici ürünlerin seçilmesi ve görüntülenmesinde kritik rol oynamaktadır. Yapılan bu çalışmada elektronik ticaret platformlarında bulunan müşterilere bir sonraki alacakları ürünlerin en doğru şekilde tavsiye edilmesi için makine öğrenmesi algoritmaları kullanılmış sonuçlar karşılaştırılmıştır. Tekil değer ayrışımı (Singular value decomposition-SVD) yönteminin daha başarılı sonuçlar elde ettiği gösterilmiştir.

Keywords

• Uzun kuyruk ürünler
• tekil değer ayrışımı
• en yakın komşuluk
• matris ayrıştırma
• sonraki öğe önerisi

References

1. [1] Oestreicher-Singer, G., and Sundararajan, A. (2012). Recommendation networks and the long tail of electronic commerce. Mis quarterly, pp. 65-83.
2. [2] Shi, L. (2013). Trading-off among accuracy, similarity, diversity, and long-tail: a graph-based recommendation approach. In Proceedings of the ACM Conference on Recommender Systems (pp. 57-64).
3. [3] Yin, H., Cui, B., Li, J., Yao, J., and Chen, C. (2012). Challenging the long tail commendation. arXiv preprint arXiv:1205.6700.
4. [4] Brynjolfsson, E., Hu, Y., and Simester, D. (2011). Goodbye pareto principle, hello long tail: The effect of search costs on the concentration of product sales. Management Science, 57(8), 1373- 1386.
5. [5] Alshammari, G., Jorro-Aragoneses, J. L., Polatidis, N., Kapetanakis, S., Pimenidis, E., and Petridis, M. (2019). A switching multi-level method for the long tail recommendation problem. Journal of Intelligent and Fuzzy Systems, 37(6), pp. 7189-7198.
6. [6] Wang, S., Gong, M., Li, H., and Yang, J. (2016). Multi-objective optimization for long tail recommendation. Knowledge-Based Systems, 104, pp. 145-155.
7. [7] Hu, X., Zhang, C., Wu, M., and Zeng, Y. (2017). Research on long tail recommendation algorithm. In IOP Conference Series: Materials Science and Engineering, vol. 261, no. 1.
8. [8] Pandey, A. K., and Ankayarkanni, B. (2020). Recommending e-commerce products on cold start and long tail using transaction data. Int. Conference on Trends in Electronics and Informatics, pp. 661-663.

9. [9] Abdollahpouri, H., Burke, R., & Mobasher, B. (2018). Popularity-aware item weighting for long-tail recommendation. arXiv preprint :1802.05382.
10. [10] Wang, Y., Wang, J., & Li, L. (2018). Enhancing Long Tail Recommendation Based on User's Experience Evolution. Int. Conference on Computer Supported Cooperative Work in Design, pp. 25-30.
11. [11] Zhou, W., and Duan, W. (2012). Online user reviews, product variety, and the long tail: An empirical investigation on online software downloads. Electronic Commerce Research and Applications, 11 (3), pp. 275-289.
12. [12] Luke, A., Johnson, J., and Ng, Y. K. (2018, November). Recommending long-tail items using extended tripartite graphs. International Conference on Big Knowledge, pp. 123-130.
13. [13] Agarwal, P., Sreepada, R. S., and Patra, B. K. (2019). A hybrid framework for improving diversity and long tail items in recommendations. Int. Conference on Pattern Recognition and Machine Intelligence, pp. 285-293.
14. [14] De Sousa Silva, D. V., De Oliveira, A. C., Almeida, F., and Durão, F. A. (2020). Exploiting Graph Similarities with Clustering to Improve Long Tail Items Recommendations. Brazilian Symposium on Multimedia and the Web, pp. 193-200.
15. [15] Cai, Y., Cui, Z., Wu, S., Lei, Z., and Ma, X. (2021). Represent Items by Items: An Enhanced Representation of the Target Item for Recommendation. ArXiv preprint arXiv:2104.12483.
16. [16] Achary, N. S., and Patra, B. K. (2021). Graph Based Hybrid Approach for Long-Tail Item Recommendation in Collaborative Filtering. ACM IKDD CODS and COMAD (pp. 426-426).
17. [17] Chen, X., Pan, Y., and Luo, B. (2020). Research on power-law distribution of long-tail data and its application to tourism recommendation. Industrial Management and Data Systems.
18. [18] Zencirli, A., Çetin, H., Tuğ, N., and Ensari, T. (2021). Deep Learning Classification of Location Oriented Recommendation System for Low-Sale Products. International Congress on Human-Computer Interaction, Optimization and Robotic Applications, pp. 1-4.
19. [19] Meenakshi, M. (2019). A Novel Approach Web Services Based Long Tail Web Services Using Deep Neural Network. Int. Conference on Issues and Challenges in Intelligent Computing Techniques, v1, pp 1-9.
20. [20] Le, D. D., and Lauw, H. (2021). Efficient Retrieval of Matrix Factorization-Based Top-k Recommendations: A Survey of Recent Approaches. Journal of Artificial Intelligence Research, 70, pp. 1441-1479.
21. [21] Cui, Z., Zhao, P., Hu, Z., Cai, X., Zhang, W., and Chen, J. (2021). An improved matrix factorization based model for many-objective optimization recommendations. Information Sciences, 579, pp. 1-14.
22. [22] Gabbolini, G., D'Amico, E., Bernardis, C., and Cremonesi, P. (2021). On the instability of embeddings for recommender systems: the case of Matrix Factorization. In Proceedings of the 36th Annual ACM Symposium on Applied Computing (pp. 1363-1370).
23. [23] Park, Y. J., and Tuzhilin, A. (2008). The long tail of recommender systems and how to leverage it. ACM conference on Recommender systems, pp. 11-18.
24. [24] Sundaresan, N. (2011). Recommender systems at the long tail. ACM conference on Recommender systems, pp. 1-6.
25. [25] Behera, R. K., Gunasekaran, A., Gupta, S., Kamboj, S., and Bala, P. K. (2020). Personalized digital marketing recommender engine. Journal of Retailing and Consumer Services, 53.
26. [26] Pratama, B. Y., Budi, I., and Yuliawati, A. (2020). Product Recommendation in Offline Retail Industry by using Collaborative Filtering. International Journal of Advanced Computer Science and Applications, 11(9), pp. 635-643.
27. [27] Valcarce, D., Parapar, J., and Barreiro, A. (2016). Item-based relevance modelling of recommendations for getting rid of long tail products. Knowledge-Based Systems, 103, pp. 41-51.
28. [28] Zhang, F., Lu, Y., Chen, J., Liu, S., and Ling, Z. (2017). Robust collaborative filtering based on non-negative matrix factorization and R1-norm. Knowledge-based systems, 118, pp. 177-190.
29. [29] Olmedilla, M., Martínez-Torres, M. R., and Toral, S. L. (2019). The superhit effect and long tail phenomenon in the context of electronic word of mouth. Decision Support Systems, 125, pp. 113-120.
30. [30] L. Zhaoyang, Matris Ayrışımı, Yüksek Lisans tezi, 2006.