Research on Financial Recommendation Algorithm Based on User Interest Evolution and Big Data

This article aims to study the recommendation algorithm of financial wealth management products based on the Internet. Through the item information of fund wealth management products, it mainly includes wealth management products issued by banks and fund products issued by fund companies as well as user behavior information, combined with traditional collaborative filtering algorithms to obtain a recommendation list, and then add a time series model, consider the seven-day annual interest rate and other time-sensitive attribute factors to improve the accuracy of the recommendation, and finally introduce the current more cutting-edge DIEN model algorithm to get the recommendation result, which integrates the sequence The model is used to simulate the change process of user interest over time, and the attention mechanism is introduced to improve the recommendation effect, thereby providing users with better product choices.