Industrial big data multi-modal feature selection method and device, equipment and medium

The invention discloses an industrial big data multi-modal feature selection method and device, equipment and a medium, and relates to the technical field of industrial big data machine learning, and the method comprises the steps: 1, data preparation: obtaining big data in an industrial production process, carrying out the preliminary preprocessing, and dividing a processed data set into a training set and a test set; 2, setting algorithm related parameters and external archives, and initializing a population; according to the industrial big data multi-modal feature selection method and device, the equipment and the medium, the data processing module, the niche division module and the niche processing module are arranged, and the Hamming similarity is designed to be used for distinguishing different feature subsets in a discrete space, so that the problem that discrete solutions are difficult to distinguish by a traditional Euclidean distance is avoided. According to the dynamic niche strategy, the niche sca