Data sharing method based on block chain and federal learning

A data sharing and blockchain technology, applied in machine learning, digital data protection, data processing applications, etc., can solve the problems of data owner privacy leakage, data privacy hindering data sharing, and unlikely sharing of local data, etc. The effect of alleviating privacy protection issues, protecting privacy and ensuring reliability

Pending Publication Date: 2022-04-29
FUJIAN NORMAL UNIV
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, data sharing in IoT may face the following problems: First, it is difficult for each organization to establish mutual trust, so they are less likely to share reliable local data; second, data privacy has become a big obstacle to

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 sharing method based on block chain and federal learning
  • Data sharing method based on block chain and federal learning
  • Data sharing method based on block chain and federal learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0088] Please refer to Figure 1 to Figure 3 , a data sharing method based on blockchain and federated learning, including steps:

[0089] S1. Form a team of blockchain nodes that trust each other.

[0090] Among them, each team has a team leader who is responsible for receiving data sharing tasks, supervising the joint learning process in data sharing, and sending the global model with differential privacy to the task issuer.

[0091] The data nodes in step S1 may have selfish behavior. To solve this problem, an internal team management mechanism based on "mortgage-punishment" is designed, which specifically includes the following steps:

[0092] S11. Provide a preset amount of collateral when forming a team, and calculate the penalty coefficient k for bad behavior nodes:

[0093]

[0094] In the formula, v represents the total number of working rounds for the node to complete the cooperative task, p represents the number of times the node temporarily exits, and q repres...

Embodiment 2

[0128] Please refer to Figure 1 to Figure 3 , the difference between this embodiment and Embodiment 1 is that it further limits the application of privacy differences to data sharing. Specifically, considering that malicious data requesters will launch attacks, the team leader should add interference to the model, using the The differential privacy model protection method specifically includes the following steps:

[0129] Given a random algorithm G, two adjacent data sets D1 and D2 with at most one different record;

[0130] After removing two data sets in a row, calculate the probability of the random algorithm G getting the same result according to formula (7):

[0131] Pr[G(D)∈0]≤exp(ε) Pr[G(D′)∈0];

[0132] In the formula, G represents a random algorithm, ε represents a privacy budget, usually a small constant, and D represents a data set;

[0133] Calculate sensitivity:

[0134] Δf=max D，D , ||G(D)-G(G′)||;

[0135] Compute the Laplacian mechanism applied to the g...

Embodiment 3

[0139] Please refer to Figure 1 to Figure 3 , the difference between this embodiment and Embodiment 1 and Embodiment 2 is that the steps of the consensus algorithm based on node contributions are further defined, specifically:

[0140] Compute the contribution of each node based on cosine similarity:

[0141]

[0142] in Indicates the actual update gradient k of the node, Indicates the local update gradient of the kth node, Indicates the model gradient before data node k is updated, Represents the gradient of the global model;

[0143] Carry out a reward mechanism based on the contribution weight ratio;

[0144] Calculate the contribution value through the mapping function:

[0145]

[0146] Use the soft-max function to calculate the weight ratio of the node's contribution to the global model;

[0147] Calculate the soft-max function value:

[0148]

[0149] Therefore, the advantage of the consensus mechanism in this embodiment is that it can prevent the ...

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 data sharing method based on a block chain and federated learning, and the method comprises the steps: building mutually trusted block chain nodes into a team, receiving a request task, and then selecting a team meeting a credit rating requirement to respond to the request task; after the data sharing task is received, the nodes in the team meeting the credit rating requirement are used for training a verification model until the verification model reaches preset accuracy or reaches the maximum training time, and model sharing is achieved to protect privacy of a data provider; a model training process is packaged locally, a consensus algorithm based on node contribution reaches a consensus among block chain nodes, and credit reward is performed on a team meeting a credit rating requirement, so that each training process in a data sharing process is recorded to ensure that a data provider provides high-quality data; the credit reward is carried out after the consensus is reached, the credit rating can be updated in time, the reliability of the credit rating is ensured, and the privacy protection problem of data in the Internet of Things is relieved.

Description

technical field [0001] The invention relates to the technical field of Internet of Things data sharing, in particular to a data sharing method based on blockchain and federated learning. Background technique [0002] With the development of Internet technology, the Internet of Things (IoT) is widely used in various industries. Sensors are an important part of IoT and the most important data source for IoT systems. The perception data collected by a single sensor often cannot meet the needs of users. The real value of the Internet of Things lies in the comprehensive utilization and sharing of various data and information. For example, in the field of healthcare, data sharing can provide valuable health records, including treatment information and physical examination information, which can provide targeted treatment to patients. In the tourism industry, by analyzing the collected data, data sharing can accurately understand the preferences of tourists, predict future touris...

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): G06F21/62G06F21/64G06Q40/02G06N20/00
CPCG06F21/6245G06F21/64G06N20/00G06Q40/03
Inventor 范新民妙秦阳汪晓丁张灵杰林晖
Owner FUJIAN NORMAL 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