Research on E-Commerce Customer Feature Extraction Question Answering System Based on Artificial Intelligence Semantic Analysis

Abstract

In order to analyze e-commerce customer behavior and preference, a migration identification method of consumer behavior tendency is proposed. Data mining technology is adopted to mine social data in individual online we-media platforms and to mine individual personal attributes and preferences from their unconscious social language. Its methods are through the customer identification model construction related research, consumer preference identification and analysis related research, based on data mining technology of consumer preference identification and analysis, and the introduction of feature extraction method: semantic analysis. According to the data, there are 2,990 customer interest consumption forecasts, 1,836 customer social network data consumption forecasts, and 3,652 customer preference consumption forecasts. In order to screen out the main factors that have the greatest impact on consumer behavior from all kinds of consumer behavior propensity factors, the multiple step-based regression method is adopted for factor selection. Because of the large difference in the multidimensional dynamic vector, the corresponding consumer behavior tendency changes greatly, so the migration identification method of consumer behavior tendency is feasible.