Electric data can accurately reflect the operation of the national economy, and can be combined with people's livelihood, environmental protection, economic and other data to create greater social and economic value. However, for the sake of privacy and security, each electric data owner is not willing to share their own data, which makes it impossible to train a more accurate model. In order to break this data island problem, this paper uses vertical federated learning to mine the value of power data and promote data circulation, under the premise of security and compliance, and has achieved good experimental results.
|