Intelligent Dynamic Collaborative Recommendation Model of User Perceived Interest Points on E-Commerce Platform
To improve the effectiveness of user perceived interest point recommendation,design an intelligent dynamic collaborative recommendation model for user perceived interest points on e-commerce platforms.Generate interest point recommendation results based on user preferences by combining user collaborative filtering recommendation results and inter-est point popularity recommendation results.Integrate geographic distance and access time information of interest points to generate recommendation results based on interest point scenario information.Using matrix decomposition model,establish an intelligent collabo-rative recommendation model for user perceived interest points.Introducing an improved hybrid hierarchical genetic algorithm to update model parameters and obtaining the best intelligent collaborative recommendation results for user perceived interest points.By an-alyzing user activity,it is possible to determine if there is a drift in the perceived interest points within the recommendation results.If there is a drift,a recommendation model is used to recalculate the recommendation results and achieve intelligent dynamic collaborative recommendation based on user perceived interest points.Experimental results have shown that this model can effectively achieve intelligent dynamic collaborative recommendation of perceived interest points for e-commerce platform users;Under different influencing factors,the product purchase conversion rate of this model is relatively high,and the recommenda-tion effect is better;After applying this model,it can effectively reduce the user churn rate of e-commerce platforms.
e-commerce platformuser perceptionpoints of interestintelligencedynamic collaborative recommendationmatrix decomposition
NETL
NSTL
万方数据
