User Behavior Analysis of E-commerce Platforms Under Random Forest Recommendation Algorithm

The 48 survey results recently released by the China Internet Network Information Center indicate that the number of online users in China has reached 1.01 billion, of which 812 million are online shoppers, accounting for 80% of online users. Among them, the number of netizens on e-commerce platforms such as Taobao, Jingdong and Tiktok exceeds 90%. When user traffic is blocked, in order to effectively compete with large e-commerce platforms, it is necessary to have accurate marketing strategies in order to bring better service to customers. E-commerce platforms need to effectively collect and analyze a massive amount of user behavior in order to achieve optimal marketing strategies and user experience. Therefore, establishing an e-commerce platform suitable for analyzing consumer behavior is very meaningful. This article uses RF (Random Forest) as a tool to study and implement a user behavior analysis system based on online shopping platforms. The feasibility of using RF algorithm for user behavior analysis on e-commerce platforms is verified through experiments (when the number of decisions reached 150, the accuracy of RF algorithm was 0.875, higher than other algorithms).