Method for clustering and analyzing features and eliminating irrelevant features according to correlation among features

The invention discloses a method for performing clustering analysis on features and eliminating irrelevant features according to correlation among the features, which comprises the following steps of: S1, removing the irrelevant features: performing clustering analysis on the features according to the correlation among the features, sorting the features from high to low according to the correlation among the features, and removing the irrelevant features; s2, feature clustering: performing feature clustering on the feature subsets after the irrelevant features are removed; s3, selecting a specified number of features; according to the method, irrelevant features are removed based on a ReliefF algorithm, then features with high information content are used as initial centers of feature clusters in feature subsets, the features are clustered according to feature association degrees, after clustering is completed, representative features are selected from each feature cluster, and feature clustering is completed