Prediction of Furniture Consumer Behavior Based on Machine Learning and Comparative Analysis of Models

This research focuses on the field of furniture products and aims to explore the application of machine learning in consumer behavior prediction. By collecting consumer behavior data in a specific period of time from the official data set of Alibaba Tianchi Data Lab, a prediction model is constructed using a variety of machine learning algorithms (including decision trees, random forests, support vector machines and XGBoost), and the model performance is evaluated using indicators such as cross validation, accuracy, recall rate and F1 score. The results show that different algorithms have their own advantages and disadvantages in predicting consumer behavior. XGBoost model performs well in all indicators, and feature selection has a significant impact on model performance. In addition, feature importance analysis reveals the important role of purchase behavior types, popular furniture types and time points in a specific time period (such as 16:00-22:00) on the prediction performance of the model. This study provides theoretical basis and data support for furniture enterprises to optimize marketing strategies, and also emphasizes the importance of data privacy protection and model generalization capability improvement.