A Novel Forecasting Method for Large-Scale Sales Prediction Using Extreme Learning Machine

摘要

With the rise of e-commerce business, sales forecasting plays an increasingly important role, for accurate and speedy forecasting can help e-commerce companies solve all the uncertainty associated with demand and supply and reduce inventory cost. As the rapid growth in the amount of data, traditional intelligence models like Neural Networks have weakness in terms of speed. In this paper, we introduce the algorithm of ELM (extreme learning machine). In addition, we subjoin many e-commerce related indicators to increase the accuracy and reliability of prediction. In sum, the new model provides a better result both in terms of speed and accuracy. Experiments are conducted with the real sales data from an e-commerce company in China.