This invention discloses a method and
system for recommending
bank points malls based on multi-dimensional user
data analysis. The method constructs an intelligent recommendation
closed loop through a six-step core process: First, it organizes users' regular consumption data, achieving multi-dimensional collection, cleaning, labeling, and dynamic updates; second, it collects and filters spatiotemporal hotspot data of the user's region to accurately match offline
scenario needs; third, it integrates multi-channel online hotspot data, quantifying popularity and timeliness; fourth, it binds the two types of hotspot data with points mall products by weight; fifth, it calculates and dynamically adjusts the correlation between products; and finally, it generates a default recommendation
list based on the hotspot correlation and dynamically switches to related product recommendations based on real-time user interaction behavior. This invention solves the problems of single recommendation dimensions, insufficient accuracy, and delayed timeliness in existing technologies. By integrating multi-dimensional data and a dynamic feedback mechanism, it significantly improves the accuracy of points mall recommendations, user interaction experience, and points redemption conversion rate.