Sales Forecasting for New Products Using Homogeneity-Based Clustering and Ensemble Method
Open Access
- 27 January 2025
- journal article
- research article
- Published by MDPI AG in Electronics
- Vol. 14 (3) , 520
Abstract
Accurate sales forecasting for new products is critical in industries characterized by intense competition, rapid innovation, and short product life cycles, such as the smartphone market. This study proposes a data-driven framework that enhances prediction accuracy by combining homogeneity-based clustering with an ensemble learning approach. Unlike traditional methods that depend on product-specific attributes, our approach utilizes historical sales data from similar products, overcoming attribute dependency. Using K-means clustering, the training data are segmented into homogeneous groups, and tailored ensemble forecasting models are developed for each cluster by combining five machine learning models: Random Forest, Extra Tree, XGBoost, LightGBM, and TabNet. When tested on South Korean smartphone sales data, the framework achieves superior performance, with the optimal ensemble model using four clusters delivering an MAPE of 8.3309% and an RMSPE of 7.8360%, significantly outperforming traditional single-cluster models. These findings demonstrate the effectiveness of leveraging data homogeneity and ensemble methods, offering a scalable and adaptable solution for accurate sales forecasting of new products.Keywords
This publication has 56 references indexed in Scilit:
- Well googled is half done: Multimodal forecasting of new fashion product sales with image‐based google trendsJournal of Forecasting, 2024
- Data set quality in Machine Learning: Consistency measure based on Group Decision MakingApplied Soft Computing, 2021
- Forecasting demand profiles of new productsDecision Support Systems, 2020
- Online sales prediction via trend alignment-based multitask recurrent neural networksKnowledge and Information Systems, 2019
- Intelligent sales volume forecasting using Google search engine dataSoft Computing, 2019
- Hyperparameters and tuning strategies for random forestWIREs Data Mining and Knowledge Discovery, 2019
- RETRACTED ARTICLE: A novel deep learning-based multi-model ensemble method for the prediction of neuromuscular disordersNeural Computing and Applications, 2018
- Forest fire forecasting using ensemble learning approachesNeural Computing and Applications, 2018
- Sales forecasting by combining clustering and machine-learning techniques for computer retailingNeural Computing and Applications, 2016
- Extremely randomized treesMachine Learning, 2006