Customer experience perception based on federated learning

An electronic device for a service provider in a wireless communication system, the service provider capable of implementing customer experience perception with at least one other service provider in the wireless communication system through federated learning, wherein the customer experience relates to communication performance of a telecommunication service product when used, the electronic device comprises a processing circuit configured to:

• acquire information comprising identifiers of samples of local data, that have been encrypted by using a specific encryption algorithm, from the at least one other service provider, wherein the specific encryption algorithm comprises a Hash value encryption algorithm;
• encrypt identifiers of samples of local data of the service provider by using the specific encryption algorithm, so that the encrypted identifiers of samples of local data of the service provider and the encrypted identifiers of samples of local data from the at least one other service provider have same forms;
• determine identifier intersection between the encrypted identifiers of samples of local data of the service provider and the encrypted identifiers of samples of local data from the at least one other service provider as common identifiers, via identifier matching;
• determine the local data of the service provider corresponding to the common identifiers as training data for training a global model related to the customer experience perception based on the federated learning, wherein the training data comprises data related to the ‘communication performance of the telecommunication service product when used by the customer;
• perform local model training by using the determined training data to obtain intermediate information related to local model training, wherein the intermediate information includes information about a result of local model training based on the local data at the service provider corresponding to the common identifiers, and wherein the result of local model training comprises at least one of model parameters and model performance loss;
• transmit the intermediate information to a coordinator, so that the coordinator can train the global model by aggregating the intermediate information from the provider and at least one other provider;
• and
• update the local model at the service provider based on information about the trained global model from the coordinator, wherein the update of the local model at the service provider comprises adjusting parameters of the local model,
• wherein the local model training at the service provider and the global model training at the coordinator are iteratively implemented, until a specific termination condition is satisfied, wherein the specific termination condition comprises one of:
• whether an iteration number reaches a specific number;
• or
• whether a model performance loss is smaller than a threshold,
• wherein the service provider and the at least one other service provider are selected based on information about data overlapping status thereamong,
• wherein information about data overlapping status may include at least one of data ID overlapping ratio and data feature overlapping ratio, and
• wherein two service providers in the system, when at least one of data ID overlapping ratio and data feature overlapping ratio is larger than a corresponding threshold, are suitable for participating in the customer experience perception based on the federated learning.