Application of Artificial Intelligence Technology in Distributed Privacy-Preserving Clustering Mining Algorithm

Abstract

With the rapid development of information technology, large-scale data sharing is a new challenge to traditional data mining methods. How to integrate into the distributed environment and obtain accurate mining results on the basis of ensuring the confidentiality of data holders has become a new research field in the field of data mining. This paper aims to study the application of artificial intelligence technology in distributed privacy-preserving clustering mining algorithms. This paper firstly introduces the basic concepts and steps of realizing the foundation of data mining and its realization process. The advantages and disadvantages of different algorithms are compared and analyzed. Next, the concept of confidentiality is introduced, the most common method of privacy protection in data mining is revision, focusing on blockchain technology, encryption technology and peer-to-peer computer security. Finally, further investigate the stored data. Based on the K-mean clustering algorithm, combined with the distribution area, this paper adopts the overall method of homomorphic encryption to minimize the clustering on distributed sites, and regard the security of the results as an intermediate link in the communication process. As long as the cluster process runs in ciphertext mode, public encryption allows the intermediate results of the cryptographic computation process to be protected, and the algorithm can record normal cluster results on a privately secure basis. Experiments show that after the improvement of distributed privacy-preserving clustering mining, the execution time of the mining algorithm is smaller than that of the traditional mining algorithm, and the average clustering accuracy is about 85%, which is high.