Electromagnetic data labeling method and system based on man-machine hybrid intelligence

The invention relates to the technical field of electromagnetic data, and discloses an electromagnetic data labeling method and system based on man-machine hybrid intelligence, and the labeling method comprises the steps: carrying out the automatic separation and labeling of electromagnetic data through a deep clustering de-interlacing separation method, and calculating a similarity matrix between different types of electromagnetic data after separation and labeling; and finally, performing manual auditing on the label of the separated and labeled electromagnetic data by a field expert according to the similarity matrix and the label confidence coefficient, thereby realizing electromagnetic data labeling with machine labeling as a main part and expert auditing as an auxiliary part. According to the method, the problems of high manpower consumption, low labeling speed, relatively low labeling accuracy and the like in the prior art are solved. 本发明涉及电磁数据技术领域，公开了一种基于人机混合智能的电磁数据标注方法及系统，该标注方法，利用深度聚类去交错分离方法对电磁数据进行自动