Data safety collaborative calculation method and system

Abstract

The invention provides a data safety collaborative calculation method and system. The method comprises the steps: 1, extracting data features, carrying out the privacy grading of data, configuring a calculation type, formulating a calculation protocol, and carrying out the uplink of the data features, the privacy grade, the calculation type and the calculation protocol; 2, after selecting an online node, a data calculation requester initiates an HTTP request and links request information, a calculation service party verifies the information on the link, establishes a secure channel after verification, carries out data transmission and calculation, and links a calculation result; and 3, verifying the calculation result. According to the method, the block chain technology is combined, the data hash before calculation and the result after calculation are chained, verification is carried out in a regular public or zero knowledge mode, and it is guaranteed that the data of the double-sending participants are real and cannot be tampered.

Description

technical field [0001] The present invention relates to the technical field of block chains, in particular to a data security collaborative computing method and system. Background technique [0002] At present, the industry lacks a set of relatively practical normative processes and methods for multi-party data collaborative computing. Most secure collaborative computing assumes that the participants are semi-real, that is, the participants need to participate in the calculation in strict accordance with the established protocol, which is difficult to verify The actual situation of the parties involved. [0003] The participants of most schemes strongly rely on the central server to forward collaborative computing tasks. The efficiency of secure multi-party computing is low, especially in scenarios with large amounts of data or computational complexity, and it is difficult to truly implement applications, such as deep learning. , large matrix operations, etc. [0004] Pate...

AI Technical Summary

Problems solved by technology

[0002] At present, the industry lacks a set of relatively practical standardized processes and methods for multi-party data collaborative computing. Most secure collaborative computing assumes that the participants are half-real, that is, the participants must strictly follow the established protocol to participate in the calculation, which is difficult to verify The actual situation of the participants

[0003] The participants of most schemes strongly rely on the central server to forward collaborative computing tasks. The efficiency of secure multi-party computing is low, especially in scenarios with large amounts of data or computational complexity, and it is difficult to truly implement applications, such as deep learning. , large matrix operations, etc.

Images