T/ZFIDA 0004-2020 Abolished Group standards

T/ZFIDA 0004-2020 Federal Learning for Financial Industry Application Guide

Publish Date: 2021-02-24 Implement Date: 2021-02-24 For services related to genuine standard inquiry, procurement, translation, and other related services in China, please Contact Us

Price: $ 20.00

Out of stock

Basic Information

Standard Code: T/ZFIDA 0004-2020

Standard Type: Group standards

Standard Status: Abolished

is_force_gb: no

CCS Name: Finance, insurance

ICS Name: Terminology (principles and coordination and cooperation)

Publish Date: 2021-02-24

Implement Date: 2021-02-24

Scope

Main technical content: This standard defines the elements, element relationships, processes, and basic requirements of federated learning applications in the financial industry. This standard applies to the solution design and development of federated learning applications for financial institutions and third-party service providers. The process of federated learning financial applications typically includes three stages: federated preprocessing, federated model training, and federated model prediction. Before participating parties generate a federated model, they usually need to preprocess the sample and label data in advance. The main tasks of the federated preprocessing stage typically include sample alignment, feature cleaning, feature processing, and federated feature selection. After completing feature preprocessing, federated model training can be carried out to generate a federated model. Depending on the selection of federated architecture and federated algorithms, federated model training can be divided into the training of linear regression, logistic regression, trees, and neural networks models. After generating a federated model, it is deployed in the production environment for prediction. Federated prediction is strongly related to federated model training, and it also includes the prediction of regression models, tree models, and neural network models.