Research on a User Context Model Based on Data Mining

With the popularity of the Internet and under the rising number of online shoppers, e-commerce platforms are facing increasing challenges in personalizing their descriptions and information services to users. Therefore, in this paper, the context factors that have an impact on users' decisions are counted through questionnaires and derived by way of statistical analysis. Based on the coarse-grained context conclusions drawn from the statistical analysis through SPSS, a web crawler was used to crawl the Taobao users' context and behavioral dataset as the experimental dataset. The crawled data was then analyzed by K-Means. Finally, the K-Means clustering model analysis is used to propose the construction of a user context model, which is used to promote the correct understanding of personalized information needs on the platform, thereby increasing user loyalty.