Machine Learning Methods for Modeling Consumer Choice: A Synthesis of Expert Judgments and Neural Network Analysis

The article explores modern machine learning methods used for modeling consumer choice, including artificial neural networks and approaches to synthesizing expert judgments. Particular attention is paid to the combination of neural models with classical decision analysis approaches to improve prediction accuracy. The paper provides an overview of neural network architectures applied to consumer preference modeling and discusses hybrid method cases in e-commerce. The effectiveness of integrating expert knowledge into machine learning structures is substantiated.