Analysis and research of retail customer consumption behavior based on support vector machine

In recent years, with rapid economic development, customers' consumption levels have significantly increased and surpassed previous levels. This has placed enormous pressure on enterprise customer relationship management systems, especially for large-scale retail businesses. Faced with expanding data resources and massive volumes of static customer information and dynamic transactional data, traditional database technologies are no longer capable of meeting the analytical needs of modern requirements. Discovering the rules, relationships, and valuable insights hidden within vast amounts of data has become a focal point and challenge for businesses' future research. With the continuous improvement and introduction of data mining methods such as Support Vector Machines (SVM) and Decision Trees, they can provide certain assistance in this regard. This article introduces Support Vector Machines and applies them to the analysis and research of customer consumption behavior in the retail industry. Firstly, a Support Vector Machine-based model, known as the DCSS model, is proposed and constructed for analyzing customer consumption behavior. The model utilizes a cloud-like processing mapping mechanism and non-deterministic reasoning, combined with the probability distribution function of SVMDT, to apply an improved Sequential Minimal Optimization (SMO) algorithm for generating classification results on customer consumption data. Subsequently, through the analysis and research of instances, the SMO algorithm based on the DCSS model is compared with other algorithms. The results demonstrate that the efficiency and accuracy of the algorithm in this article are relatively high, making it suitable for application in customer classification scenarios in the retail industry.