Design of user behavior analysis model of e-commerce website based on Spark

In the context of the era of big data, the demand for website user behavior analysis is increasing. This paper proposes a real-time analysis system for user behavior of e-commerce websites. The system is based on the streaming data processing framework-Spark Streaming. The system combines HBase database, log collection tool Flume, and distributed publish and subscribe messaging system Kafka and relevant technologies. The system includes real-time data collection and preprocessing module, real-time analysis module, and visual display module. E-commerce website users have four common behaviors: click, bookmark, add to shopping cart and make payment. This article is based on their performance in different commodity categories and different time periods. The real-time analysis system analyzes the overall distribution and trend of user behavior. The conclusion is that this article provides a strong basis for helping operators improve the service and maintenance of the website.