Utilizing Internet Big Data and Machine Learning for Product Demand Forecasting and Analysis of Its Economic Benefits

In the context of digitalization and big data-driven advancements, the accuracy of demand forecasting in supply chain management has become a key competitive factor for businesses.This paper introduces a hybrid model combining Graph Convolutional Networks (GCN), Long Short-Term Memory networks (LSTM), and attention mechanisms, which enhances forecasting performance by integrating internet big data.The model extracts key information from multiple data sources, uses GCN to capture complex relationships within the supply chain, and employs LSTM for processing time-series data, while the attention mechanism boosts sensitivity to critical time points and relationships, significantly improving prediction accuracy.Moreover, the model optimizes production plans and inventory management, reduces the risk of supply chain disruptions, and enhances market adaptability and competitiveness.