Short-Term Sales Forecasting Using LSTM and Prophet Based Models in E-Commerce

Öz

The accuracy of sales forecasting is crucial for e-commerce businesses to optimize inventory management, pricing decisions, marketing strategies and staff scheduling. At this point, different approaches such as statistical models, fuzzy systems, machine learning and deep learning algorithms are widely used for sales forecasting. This study investigates the performance of the deep learning based the Long-Short Term Memory (LSTM) model and the Facebook Prophet model on short-term sales forecasting. The performance of the proposed models is compared with the seasonal autoregressive integrated moving average (SARIMA) using real-life data from an e-commerce site. For the comparative analysis of the proposed forecasting models, weighted average absolute percent error (wMAPE), root mean square error (RMSE) and R-squared are selected as performance measures. The numerical results show that the LSTM model outperforms the Prophet and SARIMA models in terms of forecast accuracy for hourly sales forecasting.

Anahtar Kelimeler

• Sales Forecasting
• E-commerce
• LSTM
• Prophet

E-Ticarette LSTM ve Prophet Esaslı Modeller Kullanarak Kısa Dönemli Satış Tahmini

Öz

Satış tahmininin doğruluğu, e-ticaret işletmelerinin envanter yönetimini, fiyatlandırma kararlarını, pazarlama stratejilerini ve personel planlamasını en iyilemesi için çok önemlidir. Bu noktada, satış tahmini için istatistiksel modeller, bulanık sistemler, makine öğrenmesi ve derin öğrenme algoritmaları gibi farklı yaklaşımlar yaygın olarak kullanılmaktadır. Bu çalışma, derin öğrenme tabanlı Uzun-Kısa Süreli Bellek (LSTM) modeli ve Facebook Prophet modelinin kısa vadeli satış tahmini üzerindeki performansını incelemektedir. Önerilen modellerin performansı, bir e-ticaret sitesinden alınan gerçek hayat verileri kullanılarak mevsimsel otoregresif bütünleşik hareketli ortalama (SARIMA) ile karşılaştırılmıştır. Önerilen tahmin modellerinin karşılaştırmalı analizi için, performans ölçütleri olarak ağırlıklı ortalama mutlak yüzde hata (wMAPE), hata kareleri ortalamasının karekökü (RMSE) ve R-kare seçilmiştir. Sayısal sonuçlar, LSTM modelinin saatlik satış tahmini için tahmin doğruluğu açısından Prophet ve SARIMA modellerinden daha iyi performans gösterdiğini göstermiştir.

Anahtar Kelimeler

• Satış tahmini
• E-ticaret
• LSTM
• Prophet

Kaynakça

1. Arunraj, N. S., & Ahrens, D. (2015). A hybrid seasonal autoregressive integrated moving average and quantile regression for daily food sales forecasting. google scholar
2. International Journal of Production Economics, 170, 321-335. https://doi.org/10.1016Zj.ijpe.2015.09.039 google scholar
3. Bandara, K., Shi, P., Bergmeir, C., Hewamalage, H., Tran, Q., & Seaman, B. (2019). Sales Demand Forecast in E-commerce Using a Long Short-Term Memory Neural Network Methodology. In T. Gedeon, K. W. Wong, & M. Lee (Eds.), Neural Information Processing (pp. 462-474). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-36718-3_39 google scholar
4. Chandriah, K. K., & Naraganahalli, R. V. (2021). RNN / LSTM with modified Adam optimizer in deep learning approach for automobile spare parts demand forecasting. Multimedia Tools and Applications, 80(17), 26145-26159. https://doi.org/10.1007/s11042-021-10913-0 google scholar
5. Chang, P.-C., Liu, C.-H., & Fan, C.-Y. (2009). Data clustering and fuzzy neural network for sales forecasting: A case study in printed circuit board industry. Knowledge-Based Systems, 22(5), 344-355. https://doi.org/10.1016/j.knosys.2009.02.005 google scholar
6. Choi, T.-M., Hui, C.-L., Liu, N., Ng, S.-F., & Yu, Y. (2014). Fast fashion sales forecasting with limited data and time. Decision Support Systems, 59, 84-92. https://doi.org/10.1016/j.dss.2013.10.008 google scholar
7. Delasalles, E., Lamprier, S., & Denoyer, L. (2019). Dynamic Neural Language Models. In T. Gedeon, K. W. Wong, & M. google scholar
8. Lee (Eds.), Neural Information Processing (pp. 282-294). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-36718-3_24 google scholar

9. Ensafi, Y., Amin, S. H., Zhang, G., & Shah, B. (2022). Time-series forecasting of seasonal items sales using machine learning - A comparative analysis. International Journal ofInformation Management Data Insights, 2(1), 100058. https://doi.org/10.1016/j.jjimei.2022.100058 google scholar
10. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Sequence Modeling: Recurrent and Recursive Nets. In Deep learning (pp. 373-420). Cambridge, Massachusetts: The MIT Press. google scholar
11. James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). Linear Regression. In An introduction to statistical learning: With applications in R (pp. 59-128). New York: Springer. google scholar
12. Ji, S., Wang, X., Zhao, W., & Guo, D. (2019). An Application of a Three-Stage XGBoost-Based Model to Sales Forecasting of a Cross-Border E-Commerce Enterprise. Mathematical Problems in Engineering, 2019, 1-15. https://doi.org/10.1155/2019/8503252 google scholar
13. Jing, X., & Lewis, M. (2011). Stockouts in Online Retailing. Journal ofMarketing Research, 48(2), 342-354. https://doi.org/10.1509/jmkr.48.2.342 google scholar
14. Loureiro, A. L. D., Miguéis, V. L., & da Silva, L. F. M. (2018). Exploring the use of deep neural networks for sales forecasting in fashion retail. Decision Support Systems, 114, 81-93. https://doi.org/10.1016/j.dss.2018.08.010 google scholar
15. Martínez, F., Charte, F., Frías, M. P., & Martínez-Rodríguez, A. M. (2022). Strategies for time series forecasting with generalized regression neural networks. Neurocomputing, 491, 509-521. https://doi.org/10.1016/j.neucom.2021.12.028 google scholar
16. Punia, S., Nikolopoulos, K., Singh, S. P., Madaan, J. K., & Litsiou, K. (2020). Deep learning with long short-term memory networks and random forests for demand forecasting in multi-channel retail. International Journal ofProduction Research, 58(16), 4964-4979. https://doi.org/10.1080/00207543 .2020.1735666 google scholar
17. Sun, Z.-L., Choi, T.-M., Au, K.-F., & Yu, Y. (2008). Sales forecasting using extreme learning machine with applications in fashion retailing. Decision Support Systems, 46(1), 411-419. https://doi.org/10.1016/j.dss.2008.07.009 google scholar
18. Taylor, S. J., & Letham, B. (2017). Forecasting at scale (No. e3190v2). PeerJ Inc. https://doi.org/10.7287/peerj.preprints.3190v2 google scholar
19. Weytjens, H., Lohmann, E., & Kleinsteuber, M. (2021). Cash flow prediction: MLP and LSTM compared to ARIMA and google scholar
20. Prophet. Electronic Commerce Research, 21(2), 371-391. https://doi.org/10.1007/s10660-019-09362-7 google scholar
21. Yu, Q., Wang, K., Strandhagen, J. O., & Wang, Y. (2018). Application of Long Short-Term Memory Neural Network to google scholar
22. Sales Forecasting in Retail—A Case Study. In K. Wang, Y. Wang, J. O. Strandhagen, & T. Yu (Eds.), Advanced Manufacturing and Automation VII (pp.11-17). Singapore: Springer. https://doi.org/10.1007/978-981-10-5768-7_2 google scholar
23. Yu, Y., Choi, T.-M., & Hui, C.-L. (2011). An intelligent fast sales forecasting model for fashion products. Expert Systems with Applications, 38(6), 7373-7379. https://doi.org/10.1016/j.eswa.2010.12.089 google scholar
24. Zhang, G. P. (2003). Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159-175. https://doi.org/10.1016/ S0925-2312(01)00702-0 google scholar
25. Zhao, K., & Wang, C. (2017, August 26). Sales Forecast in E-commerce using Convolutional Neural Network. arXiv. https://doi.org/10.48550/arXiv.1708.07946 google scholar
26. Zohdi, M., Rafiee, M., Kayvanfar, V., & Salamiraad, A. (2022). Demand forecasting based machine learning algorithms on customer information: An applied approach. International Journal ofInformation Technology, 14(4), 1937-1947. https://doi.org/10.1007/s41870-022-00875-3 google scholar