The research and experimental analysis of machine learning methods in e-commerce tasks

Author(s) Collection number Pages Download abstract Download full text
Havrysh B. M., Tkachenko R. O., Vitynskyi P. B. № 1 (58) 62-70 Image Image

The development of commercial activity by means of information systems provides many advantages for consumers of services and for companies that make such pro­posals. Exploring new markets for goods and services by internet companies provides great opportunities to consumers for analyzing and buying new products. The rapid development of computer technologies and information systems allows customers to buy the product even on the phone. However, the part of retail sales using such systems is still very small. A traditional e-commerce model based on finding a product or service on the Internet is becoming a bottleneck, impeding further company development. Instead, it can be replaced by personalized systems that give many benefits.

The problem of big data analytics in personalized systems of electronic commerce has been considered. An overview of existing methods of machine learning (Random Forest, AdaBoost, Multilayer Perceptron, SGTM neural-like structure, Linear Regression using Gradient Descent, General Regression Neural Network, Regressor Based on the Support Vectors Machine) for solving regression problems in such systems has been made. The advantages and disadvantages of the considered methods and algorithms have been presented. The methods of machine learning have been used for forecasting the number of consumer expenses of the retail store. The accuracy of the forecasting and the time of training procedures of all considered methods has been experimentally determined. Despite the high temporal characteristics of the work, methods based on the Support Vectors Machine and the Stochastic Gradient Descent do not provide sufficient precision. The efficiency of neural network methods is also not satisfactory in terms of accuracy and training time. It has been established that the ensemble methods provide the highest accuracy with satisfactory time characteristics of the training procedures for a given problem.

Keywords: ensemble, e-commerce, machine learning, forecasting, random forest, regression task, artificial intelligence.

doi: 10.32403/1998-6912-2019-1-58-62-70

  • Lee, J. K. (2000). Artificial Intelligence Applications in Electronic Commerce. PRICAI 2000 Topics in Artificial Intelligence. Springer, Berlin, Heidelberg, 4–42 (in English).
  • Cheng, D. (2011). The Research of Personalization E-Commerce Model Based on Data Mining. International Conference on Management and Service Science, 1–43 (in English).
  • Vysotska, V., & Chyrun, L. (2014). Conceptual model of electronic content commerce sys­tems. Radio Electronics, Computer Science, Control, 1, 46–54 (in English).
  • Raghavan, N-R. S. (2005). Data mining in e-commerce: A survey. Sadhana, 30, 2–3, 275–289 (in English).
  • Karat, C-M, Blom, J. O., & Karat, J. (2004). Designing Personalized User Experiences in eCommerce. Springer: Human–Computer Interaction Series. Springer, Netherlands, 348 (in English).
  • Smith, M., Wenerstrom, B., Giraud-Carrier, C., Lawyer, S., & Liu, W. (2007). Per­sona­­li­zing E-Commerce with Data Mining. In: Lu J., Zhang G., Ruan D. (eds) E-Service In­telligence. Studies in Computational Intelligence. Springer, Heidelberg, 37, 273–286 (in En­glish).
  • Tkachenko, R., Duriagina, Z., & Lemishka, I. et al. (2018). Development of machine learning method of titanium alloy properties identification in additive technologies. Eastern-European Journal of Enterprise Technologies, 3, 23–31 (in English).
  • Joshi, R., Gupte, R., & Saravanan, P. (2018). A Random Forest Approach for Predicting On­line Buying Behavior of Indian Customers. Theoretical Economics Letters, 08, 448 (in English).
  • Wu, X., & Meng, S. (2016). E-commerce customer churn prediction based on improved SMOTE and AdaBoost. 13th International Conference on Service Systems and Service Ma­nagement (ICSSSM). Kunming, 1–5 (in English).
  • Cao, Y., Miao, Q-C., Liu, J-C., & Gao, L. (2013). Advance and Prospects of AdaBoost Algorithm. Acta Automatica Sinica, 39, 6, 745–758 (in English).
  • Alomair, O. A., & Garrouch, A. A. (2016). A general regression neural network model offers reliable prediction of CO2 minimum miscibility pressure. Journal Petrol Explor Prod Technol, 6, 351–365 (in English).
  • Tkachenko, R., & Izonin, I. (2019). Model and Principles for the Implementation of Neural-Like Structures Based on Geometric Data Transformations. In: Hu Z, Petoukhov S, Dychka I, He M (eds) Advances in Computer Science for Engineering and Education. Springer International Publishing, Cham, 578–587 (in English).
  • Izonin, I., & Trostianchyn, A. et al. (2018). The Combined Use of the Wiener Polynomial and SVM for Material Classification Task in Medical Implants Production. International Journal of Intelligent Systems and Applications, 10:40–47 (in English).
  • Tepla, T. L., Izonin, I. V., & Duriagina, Z. A. et al. (2018). Alloys selection based on the supervised learning technique for design of biocompatible medical materials. Archives of Materials Science and Engineering, 1, 32–40 (in English).
  • Black Friday Bonanza. (2019). Retrieved from (30.05.2019) (in English).
  • EDA + REGRESSION + CLASSIFICATION FROM SCRATCH. (2019). Retrieved from (30.05.2019) (in English).