DEMAND FORECASTING IN E-COMMERCE BASED ON THE INTEGRATION OF QUANTITATIVE AND QUALITATIVE DATA: ISSUES AND PROSPECTS OF BIG DATA ANALYSIS
Back to articles of current journal| PDF: |  |
Author(s): | Gertsenberger E. A., |
| Number of journal: | 2(71) | Date: | June 2025 |
| Annotation: | In the context of rapid digitalization of com- merce and the growing volume of both structured and unstructured data, the implementation of Big Data analy- sis technologies in demand forecasting and strategic man- agement processes in e-commerce is becoming increasing- ly relevant. This study aims to identify current challenges and assess the prospects of applying analytical methods using the online clothing sales market as a case study. The objective of the research is to develop a methodological approach for the comprehensive analysis of heterogeneous data in order to improve forecasting accuracy and optimize business processes. The research methodology includes the automated collec- tion of data from both internal sources (sales volume, inven- tory levels, conversion rates at various stages) and external sources (sales volumes of similar products on marketplaces, trends, seasonality, brand mentions, and customer reviews), followed by data cleaning, normalization, and structuring. In the forecasting section, time series models — Prophet and SARIMA — as well as neural network approaches are used to account for both seasonal and behavioral factors of demand. The results are validated using real data from one product category, and forecasted values are compared with actual indicators. The practical significance of the study lies in the development of an algorithm that enables e-commerce participants to improve planning accuracy, adapt product assortments to current consumer needs, and reduce costs associated with asset freeze in inventory. The conclusion outlines future research directions, including the development of methods for eliminating data distortion, automatic filtering of unreliable information, integration of multimodal sources, and improving model adaptability to changing market conditions. | ||
| Keywords: | Big Data analysis, business analytics, demand forecasting, consumer preference analysis, sales analytics, fea- tures of online sales, e-commerce, integration of heterogeneous data, neural networks, automated data collection |
||
| For citation: | Gertsenberger E. A. Demand forecasting in e-commerce based on the integration of quantitative and qualitative data: issues and prospects of Big Data analysis. Biznes. Obrazovanie. Pravo = Business. Education. Law. 2025;2(71):141—146. DOI: 10.25683/VOLBI.2025.71.1318. |
||