Differential privacy protection-oriented k-means clustering method adopting
A differential privacy and clustering method technology, applied in the field of information security, can solve problems such as reducing the availability of clustering results, avoid the blindness of k value and initial point sensitivity, improve usability, and reduce the number of iterations.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Publication Date
- 2018-07-13
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The invention relates to a privacy protection and clustering method, in particular to a differential privacy protection-oriented k-means clustering method, which belongs to the technical field of information security. Background technique
[0002] With the rapid development of cloud computing and big data, data mining technology has made great progress in some in-depth research and applications. As one of the important methods of data mining, clustering algorithm can mine implicit and unknown knowledge and rules, and has important potential value in business decision-making of a large amount of relevant data. But at the same time, a large amount of information disclosure of sensitive information brings immeasurable threats and losses to users. Therefore, how to protect data privacy in the process of cluster analysis has become a hot issue in the field of data mining and data privacy protection. With the proposal and development of privacy protection ...
Examples
Embodiment Construction
[0036] The implementation of the technical solution of the present invention will be described in further detail below in conjunction with the accompanying drawings. It should be understood that these examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention. After reading the present invention, those skilled in the art will understand the present invention Modifications in various equivalent forms fall within the scope defined by the appended claims of the present application.
[0037] A differential privacy-oriented k-means clustering method of the present invention uses the result of the k-means++ algorithm as an input value, and then performs a series of non-local "jumps" alternately with the traditional k-means algorithm to improve the initial The selection of the center point, and using the differential privacy protection Laplace mechanism, increases the appropriate random noise that satisfies a specific di...