Privacy protection k-means clustering method and device, medium and terminal

Abstract

The invention belongs to the technical field of information privacy protection, and discloses a privacy protection k-means clustering method and device, a medium and a terminal, and the privacy protection k-means clustering method comprises the steps: collecting and encrypting original data, processing the encrypted data through two servers, classifying the data through a K-means algorithm, and grouping similar data according to distance measurement. The privacy information of the user can be well protected in the whole K-means clustering algorithm process, and the invention is friendly in the two aspects of calculation cost and time. The privacy protection k-means clustering method provided by the invention has the advantages of user friendliness, safety, correctness and effectiveness. The efficient privacy protection k-means clustering method provided by the invention is efficient and has linear complexity, and correctness and efficiency are not affected under the condition that the problem of user information privacy is solved.

Description

technical field [0001] The invention belongs to the technical field of privacy protection, and in particular relates to a privacy protection k-means clustering method, equipment, medium and terminal. Background technique [0002] Present: Machine learning is an important and a hot topic in the current environment. The advent of machine learning has led to breakthroughs in solving numerous problems in various fields, such as recommendation services, spam filtering, web search engines, fraud detection, stock market analysis, and authentication technology. While recent technologies enable more efficient storage and computation of big data, protecting the combined data from different sources remains a major challenge. [0003] However, machine learning needs to rely on a large amount of training data. The original data is messy. At the same time, the existing K-means algorithm is used to classify the data directly in the plaintext, which will lead to the leakage of the user's p...

AI Technical Summary

Problems solved by technology

[0005] Through the above analysis, the problems and defects of the existing technology are: the existing clustering method makes the user's private information and classification results extremely easy to be leaked, and the security is not enough. At the same time, the time cost of the existing clustering method based on privacy protection Large, low clustering efficiency, and inaccurate clustering results

Images