FEATURE GROUPING NORMALIZATION METHOD FOR COGNITIVE STATE RECOGNITION

摘要

A feature grouping normalization method for cognitive state recognition, which relates to the feature normalization problem in the field of pattern recognition. The method comprises the steps of: (1) performing feature data grouping; (2) optionally selecting a normalization function, and calculating parameters of normalization functions corresponding to all groups; (3) constructing a grouping normalization function, substituting the parameters of the normalization functions corresponding to all the groups into the functions of the groups, and obtaining normalization mapping relations of all the groups; and (4) performing grouping normalization processing, using the corresponding normalization function to perform feature data conversion for each group, and ending feature normalization. According to an overall feature normalization method, only the diversity problem of data distribution among features can be solved, and the problem of the overlarge distribution difference of feature internal data cannot be solved. According to the grouping normalization method provided in the present invention, the advantage of the overall feature normalization method is maintained, the problem of the overlarge distribution scale in the feature data is solved, and the classification accuracy is improved. The feature grouping normalization method provided in the present invention has strong robustness.