A binary classification method for EEG signal feature classification based on analytic hierarchy process

Abstract

The present invention relates to a binary classification method for classification of EEG signal features based on hierarchical analysis. The specific steps of the method include: collecting motor imagery EEG signals by a brain-computer interface system; performing format conversion and data preprocessing; using feature extraction algorithms to separate Extract the EEG signal features of different time periods and frequency segments; divide the extracted EEG signals and their corresponding markers into a training set and a test set according to an appropriate ratio; use a binary classification algorithm to obtain a predictive classification model according to the training set; Use the predictive classification model to calculate the test set classification results of the test set, and obtain the correct rate of the results in different time periods and frequency periods; compare them with the actual results to construct a judgment matrix; conduct a consistency check on the judgment matrix; determine the difference The weight of the time segment or different frequency segment judgment matrix; the weight is used to weight the training results of different time segments or frequency segments to obtain the final result and its corresponding credibility.

Description

technical field [0001] The invention belongs to the technical field of electroencephalogram signal processing, and in particular relates to a binary classification method for classifying electroencephalogram signal features based on hierarchical analysis. Background technique [0002] Brain Computer Interface (BCI) refers to a device that enables people to communicate or control the outside world without relying on the peripheral nervous system and muscles. As a new communication and control technology, brain-computer interface technology can provide language communication and environmental control means for patients with normal thinking but severe motor impairment. In addition, brain-computer interface technology is not only used to provide language communication and environmental control for patients, but also has potential application value in scientific fields such as automatic control and military cognition. In view of its huge application prospects, the brain-computer...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the above problems, the present invention overcomes the problems in the prior art in the feature classification of EEG signals that the classification basis features are single or cannot better synthesize multiple features, the classification accuracy is not high, and the reliability of the classification results is unknown. A binary classification method based on the classification of EEG signal features based on hierarchical analysis, which can improve the classification accuracy and the credibility of the classification results

Images