Answers are grounded in this paper and cite their sources.
In the reality of deepening population aging and continuous weakness of residential household consumption, it is of great practical significance to discuss how to better satisfy household retirement needs. In this paper, XGBoost is used to establish a user purchase behavior prediction model, and the hyperparameters of the model are optimized using genetic algorithm. The GA-XGBoost model is used to predict the digital consumption behavior of family aging and is divided according to different categories of family aging characteristics. The online platform dataset shows that general value home pension customers are more inclined to purchase goods with high conversion rates and less inclined to purchase goods with higher average prices of goods in the category compared to important home pension customers. The GA-XGBoost model correctly predicts the digital consumption behavior of the general value home pension customers with a rate of 0.8636. Compared to traditional machine learning such as logistic regression, decision tree methods and single XGBoost model, the prediction accuracy of GA-XGBoost model is significantly improved.
No takes yet. Share an insight, caveat, or question.
Jing An (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: