Federal learning-based sample evaluation method, federal learning-based sample evaluation device and electronic equipment

The invention discloses a federal learning-based sample evaluation method and device and electronic equipment, and relates to the field of artificial intelligence, and the evaluation method comprises the steps: carrying out the privacy intersection processing of a data requester and a data provider, and obtaining a first sample data set of the data requester and a second sample data set of the data provider; inputting the first sample data set and the second sample data set into an evaluation model, and outputting a tag value set corresponding to the data provider; and evaluating the sample based on the label value of each sample in the label value set to obtain an evaluation result. According to the method and the device, the technical problem that when the samples are evaluated on the basis of privacy intersection, for sample data with sample inclination, the intersection result is resampled according to the weight, so that weight information is leaked to the other party, and label information of the sample