With the rapid development of network, communication, and information technology, e-commerce has received widespread attention as a new business model. E-commerce has become a popular term, and e-commerce has become a popular mainstream in universities. However, as a comprehensive and free subject of art and science penetration, e-commerce puts forward more comprehensive requirements for teaching content, process and methods. Based on the above background, the purpose of this article is to study the professional practice teaching methods of e-commerce based on data mining algorithms. This article first analyzes the privacy issues caused by decentralized data mining under the condition of rapid development of e-commerce. How to build a more accurate model to complete data mining tasks without revealing personal privacy, mainly discussing data mining privacy protection algorithms. Next, through literary research method, questionnaire survey method, case research method, and interview method, explore the practical teaching method of e-commerce professional based on data mining algorithm. The role is to study the problems and solutions that blended learning needs to solve in specific applications. The experimental results show that, compared with the previous guidance method, the new teaching method can improve students’ initiative, enthusiasm and creativity by about 10% in the learning process.