Research On Clustering Algorithm Based On Homomorphic Encryption And Its Application In Precision Marketing

In the context of the gradual development of the Internet economy,the online dividend of the mobile Internet has gradually ebbed,and the opportunity for new entrances for enterprises lies in transforming the strategic focus into in-depth mining and systematic extraction and utilization of offline traffic demand.Digital drive and intelligent operation are gradually reflected in the real consumption economy,and the value of big data for offline consumption is also becoming increasingly prominent.The combination and application of big data and machine learning in the offline consumer market will continuously improve the scientificity of marketing strategy formulation.The use of machine learning can make statistics and analysis of users' consumption behaviors,and classify customers according to users' consumption behaviors,so as to realize personalized push of marketing content.The training of machine learning models requires a large amount of data.Driven by economic interests,consumers' personal information and consumption records are illegally collected,stored,used and even sold,which not only violates personal privacy,but also brings personal and property losses to consumers.Out of the consideration of consumer information security,how to safely collect and use this data is an urgent problem to be solved.In addition,how to propose a privacy protection model suitable for consumer data application scenarios under the premise of meeting data privacy,security and regulatory requirements,so that the model can formulate precision marketing strategies more scientifically based on a large amount of consumer data,which is the key to the current competition for offline market.In view of the above problems,this article conducts research from the following aspects:(1)Analyze the current domestic and foreign research on privacy protection of machine learning,and conduct research on typical machine learning privacy protection technologies.Then,introduce Vector Homomorphic Encryption into data sharing to ensure data privacy and security in the process of machine learning.(2)Analyze the existing problems in the current consumption data sharing,while ensuring data security,introduce the K-means algorithm into customer segmentation to lay the foundation for the realization of precision marketing.And from the perspective of practical applications,design precision marketing programs based on consumption data and homomorphic encryption algorithms.(3)In view of the limitations of the original homomorphic encryption algorithm,the range of homomorphic encryption processing data is extended to rational numbers.Then rewrite the K-means algorithm to solve the problem that the algorithm cannot cluster on the large integer ciphertext domain.In order to solve the problem that the customer category is difficult to locate after clustering,the K-means algorithm is improved to distinguish the category of customers by changing the generation method of the initial center point,so that the algorithm is more suitable for the scheme designed in this paper.(4)According to the design of the precision marketing plan,a consumption data analysis system based on homomorphic encryption is implemented,and the system functions,modules and databases are designed in detail,so that the plan proposed in this article can be realized in the system.The main functions of the system are tested to ensure that the system achieves the design goals of precision marketing based on consumption data and homomorphic encryption algorithms.