Unlock instant, AI-driven research and patent intelligence for your innovation.

Data processing method and device based on privacy protection

A privacy protection and data processing technology, which is applied in the fields of electronic digital data processing, digital data protection, computer security devices, etc., and can solve the problems of low parallelism, many times of communication and many times of compilation.

Active Publication Date: 2021-01-29
BEIJING REALAI TECH CO LTD
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the process of transforming the XGBoost algorithm into a data flow graph, the programming model needs to be called multiple times, resulting in problems such as high compilation times, high communication times, and low parallelism.

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
  • Data processing method and device based on privacy protection
  • Data processing method and device based on privacy protection
  • Data processing method and device based on privacy protection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] figure 2 A flow chart of a data processing method based on privacy protection provided by an embodiment of the present disclosure, the method includes the following steps S202 to S210:

[0033] Step S202, acquiring a machine learning algorithm to be trained, and the machine learning algorithm is an algorithm including dynamic multi-party interaction and static multi-party interaction. The machine learning algorithm in this embodiment takes the XGBoost algorithm as an example.

[0034] The dynamic multi-party interaction is: the calculation process of the interaction between the first participant and the second participant contains dynamic instructions such as supporting dynamic instructions and circular dynamic instructions, and when the data content of the first participant and / or the second participant changes , the calculation process such as calculation times and calculation order will also change accordingly. Static multi-party interaction is: the calculation proc...

Embodiment 2

[0061] This embodiment provides a data processing device based on privacy protection, the device comprising:

[0062] Algorithm acquisition module, used to acquire the machine learning algorithm to be trained;

[0063] The parameter conversion module is used to obtain multiple sets of feature data that need to be called repeatedly from the machine learning algorithm, and convert the acquired feature data into tuple variable parameters of a composite data structure; wherein, the composite data structure includes: an array, a dictionary or a set ;

[0064] The data flow graph generation module is used to input the tuple variable parameters into the programming model, so that the programming model converts the machine learning algorithm based on the tuple variable parameters and the preset data flow graph generation tool, and obtains the data corresponding to the machine learning algorithm Flow graph; data flow graph includes a series of operators;

[0065] The segmentation sch...

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 relates to a data processing method and device based on privacy protection. The method comprises the steps of: obtaining a to-be-trained machine learning algorithm; obtaining multiple groups of feature data needing to be repeatedly called from the machine learning algorithm, and converting the obtained feature data into tuple variable parameters of composite data structures such as arrays, dictionaries or sets; inputting the tuple variable parameters into a programming model to enable the programming model to convert the machine learning algorithm based on the tuple variable parameters and a preset data flow graph generation tool to obtain a data flow graph corresponding to the machine learning algorithm, the data flow graph comprising a series of operators; segmenting the data flow graph into a plurality of sub-graphs, and scheduling the sub-graphs to a target participant for execution; and compiling the sub-graphs into a new data flow graph, and generating an instruction of each operator in the new data flow graph to obtain a privacy protection machine learning algorithm. With the data processing method and device based on privacy protection of the invention adopted, the number of compiling times can be decreased, and the degree of parallelism can be improved.

Description

technical field [0001] The present disclosure relates to the technical field of data encryption, and in particular to a data processing method and device based on privacy protection. Background technique [0002] In machine learning scenarios that require the participation of multiple parties, it is necessary for multiple parties to participate in information interaction to complete model training, and it is also necessary to ensure data security and protect data privacy. XGBoost (eXtreme Gradient Boosting, extreme gradient boosting) algorithm is an algorithm in machine learning scenarios. For the XGBoost algorithm, the data privacy protection in the machine learning process is mainly that the X data (feature data) and Y data (label data) are in two different parties, and both parties need to complete it without revealing the X data and Y data. XGBoost algorithm training. [0003] At present, the XGBoost algorithm with dynamic multi-party interaction function can be transf...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F21/62G06F21/60
CPCG06F21/602G06F21/6245
Inventor 徐世真朱晓芳倪裕芳王鲲鹏刘荔园唐家渝田天
Owner BEIJING REALAI TECH CO LTD