Research on Dynamic Market Demand Forecasting based on Machine Learning

DOI

10.2991/978-94-6463-690-1_28How to use a DOI?

Keywords

Dynamic market demand forecasting; Long short-term memory networks; Time series modeling; Machine learning

Abstract

In the contemporary market, which is characterised by rapid change, the ability to accurately forecast consumer demand is of paramount importance for businesses seeking to optimise a range of operational aspects, including inventory management, pricing strategies and supply chain operations. This work puts forth a dynamic market demand forecasting model based on the Long Short-Term Memory (LSTM) network. It employs the formidable time series modeling capacity of LSTM to discern long-term dependencies and intricate nonlinear relationships in market demand data. An adaptive training mechanism has been introduced, enabling the model to adjust its weights in real time based on new input data, thereby improving its responsiveness to market fluctuations. Furthermore, the proposed model enhances prediction accuracy and mitigates the uncertainty associated with manual parameter tuning by optimizing the automatic search mechanism for hyperparameters. The LSTM model is capable of processing multidimensional input data, including historical sales data, promotional activities, seasonal factors, and external economic indicators, in order to comprehensively understand the key factors influencing market demand. The experimental results demonstrate that the proposed model exhibits notable advantages in capturing the intricate dynamic alterations of market demand, particularly in the domain of forecasting long-term series and multidimensional influencing factors. Its predictive precision is superior to that of alternative models.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Xiang Yue
PY  - 2025
DA  - 2025/04/23
TI  - Research on Dynamic Market Demand Forecasting based on Machine Learning
BT  - Proceedings of the 2024 6th International Conference on Economic Management and Model Engineering (ICEMME 2024)
PB  - Atlantis Press
SP  - 289
EP  - 296
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6463-690-1_28
DO  - 10.2991/978-94-6463-690-1_28
ID  - Yue2025
ER  -