Decision model privacy protection publishing method based on objective function disturbance

An objective function, privacy protection technology, applied in the field of privacy protection, can solve the problems of reduced decision-making model accuracy, participation of user privacy leakage, etc.

Pending Publication Date: 2021-08-20
XIAN UNIV OF POSTS & TELECOMM
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem that the existing intelligent decision-making model will bring privacy leakage risks to participating users when it is released, and the direct application of privacy prot

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Decision model privacy protection publishing method based on objective function disturbance
  • Decision model privacy protection publishing method based on objective function disturbance
  • Decision model privacy protection publishing method based on objective function disturbance

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0059] Such as figure 1 As shown, the decision-making model privacy protection publishing method based on the objective function disturbance provided by the present invention specifically includes the following steps:

[0060] Step1 build the target function

[0061] First, obtain the decision data set D owned by each user i i , i∈{1,2,…,N}, where N is the total number of users; each user i owns a decision-making data set D containing n records i Expressed as:

[0062]

[0063] in, Indicates that user i selects an option instead of selecting the option A record of j∈{1,2,…,n}; and Both are d-dimensional vectors, each representing an option in the decision-making scenario, and d represents the dimension;

[0064] Then, define two options and The difference is V j i ,Right now

[0065]

[0066] Then, based on the Gaus...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a decision model privacy protection publishing method based on objective function disturbance, and aims to solve the problems that a privacy leakage risk is brought to participating users when an existing intelligent decision model is published, and the accuracy of the decision model is severely reduced due to direct application of a privacy protection technology. The method comprises the following steps: firstly, constructing an objective function according to user decision data in an intelligent decision scene, performing decomposition and Taylor expansion on the objective function, then calculating a sensitivity upper bound based on a privacy protection model, finally disturbing the objective function, calculating model parameters according to the disturbed objective function, so the privacy protection release of the decision model is realized. Compared with other privacy protection methods which directly disturb the parameters of the decision model, the method has the advantages that the target function of the decision model is disturbed, excessive noise is prevented from being introduced into the model parameters, and therefore the accuracy of the decision model with privacy protection is effectively improved.

Description

technical field [0001] The invention relates to a privacy protection method, in particular to a privacy protection release method for a decision model based on objective function disturbance. Background technique [0002] At present, artificial intelligence technology has gradually entered all aspects of people's production and life. Among them, the intelligent decision-making model, as an important model of artificial intelligence technology, plays an important role in recommendation systems and decision-making systems. Perception data from a large number of users in various industries supports the accurate and efficient execution of intelligent decision-making models. However, when third-party service providers collect and analyze a large amount of sensory data to provide users with corresponding intelligent decisions, it also brings unprecedented privacy leakage risks to participating users, which may lead to serious exposure of personal privacy. For example, some attack...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F21/62G06F17/15G06F17/17
CPCG06F21/6245G06F17/15G06F17/17
Inventor 王腾冯景瑜张雪锋
Owner XIAN UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap