Safe data-outsourcing machine learning and data analysis method

A data outsourcing and machine learning technology, applied in the field of information security, can solve problems such as data leakage, reduce complexity, improve computing efficiency, and solve the contradiction between security and processability

Active Publication Date: 2017-09-01
XIDIAN UNIV
View PDF7 Cites 40 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the deficiencies in the prior art above, the purpose of the present invention is to provide a safe data analysis method for outsourcing machine learning, which solves the contradiction between the security

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
  • Safe data-outsourcing machine learning and data analysis method
  • Safe data-outsourcing machine learning and data analysis method
  • Safe data-outsourcing machine learning and data analysis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] In Example 1, we use medical care as a scenario to simulate. In this scenario, the hospital is the model provider and the patient is the model user. The hypothetical model is as follows: Figure 4 shown.

[0058] First, the hospital converts its own 4-layer decision tree model into a polynomial and encrypts it. The polynomial calculation result is 1-8 corresponding to c1-c8, and the polynomial before encryption is

[0059] b0*b2*b6*8+b0*b2*(1-b6)*7+b0*(1-b2)*b5*6+b0*(1-b2)*(1-b5)*5+(1 -b0)*b1*b4*4+(1-b0)*b1*b4*3+(1-b0)*(1-b1)*b3*2+(1-b0)*(1-b1)* (1-b3)*1

[0060] After polynomial encryption is

[0061] b0*b2*b6*D522323233434223, b0*b2*(1-b6)*23343923822291EF,

[0062] b0*(1-b2)*b5*223DFD838D932BCB, b0*(1-b2)*(1-b5)*3343422323343923,

[0063] (1-b0)*b1*b4*838D9333434223D, (1-b0)*b1*b4*34342DFD83933343,

[0064] (1-b0)*(1-b1)*b3*32323FD838DD9333, (1-b0)*(1-b1)*(1-b3)*DFD832BCB3232323.

[0065] Then, the patient encodes his physiological data into a binary sequence ...

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 safe data-outsourcing machine learning and data analysis method. A random tree is transformed in a homoenergetic mode into properties of a binary tree; a trained decision tree prediction model is transformed into the binary tree, the binary tree is induced into a special polynomial of which a shape is a sum of an infinite number of multiplier items, and each piece of data in the model is encrypted by an RSA and is uploaded to a cloud platform; then data which needs to be decided is also encrypted by the RSA and is uploaded to the cloud platform; the encrypted data of the model and the encrypted data which needs to be decided are correspondingly computed by utilizing the multiplication homomorphic encryption property of the RSA to obtain a ciphertext result of each multiplier item; and the results are returned to a data user for decryption so as to obtain a decision result. By transforming the binary tree into the polynomial, a decision tree which originally can be implemented by various computations is transformed into the decision tree which can be implemented by one computation, so that the machine learning process of the decision tree can be rapidly completed by using a multiplication homomorphic algorithm, a complex degree of carrying out such decision tree machine learning algorithm on a ciphertext is greatly reduced, and computing efficiency is promoted.

Description

technical field [0001] The invention belongs to the technical field of information security, and in particular relates to a safe data analysis method for data outsourcing machine learning. Background technique [0002] With the development of science and technology, the advantages of cloud platforms (Cloud Platforms), such as huge scale, fast virtualization, strong versatility, on-demand service and very low cost, have become increasingly prominent, which makes high-speed computing and efficient storage of big data a reality. However, the cloud platform has always had huge security risks, and data leakage incidents are extremely prone to occur. In just a few years, shocking information leakage incidents have occurred frequently: Anthem, the second largest medical insurance company in the United States, lost 80 million personal information, and the resume data of 58, a well-known domestic job search platform, was sold at a low price. Personal privacy data is stored centrally...

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): H04L9/30H04L9/00H04L29/06G06N99/00G06F21/60
CPCG06F21/602G06N20/00H04L9/008H04L9/302H04L63/0442
Inventor 赵姝畅骆苑新雨郭娟娟马建峰王祥宇常益嘉马莹莹
Owner XIDIAN UNIV
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