Özet:

Brain computer interfaces (BBAs) are systems that allow the use of assistant technologies such as computer or neuroprosthetic arms by converting brain electrical activities into control commands. This study has studied the achievements that the use of filter-based identity selection methods along with different classification algorithms can bring to BBA systems. In this framework, EEG records obtained from the BBA system designed to control a neuroprosthetic device have been analyzed. Delta for the analysis of EEG records (1). 0-4 Hz), teta (4-8 Hz), alfa (8-12 Hz), beta (12-25 Hz), high-beta (25-30 Hz) and gamma (30-50 Hz) frequency bands and delta (1-4 Hz), teta (4-8 Hz), alfa1 (8-10 Hz), alfa2 (10-12 Hz), beta1 (12-15 Hz), beta2 (15-18 Hz), beta3 (18-25 Hz), gamma1 (30-35 Hz), gamma2 (35-40 Hz), gamma3 (40-50 Hz) low-frequency bands have been removed band power properties. The two obtained data sets are classified by applying the proprietary choice and applying the proprietary choice. In the study, together with a total of 10 filter-based proprietary selection methods, linear distinctive analysis, racial forests, decision trees and support vector machines classification algorithms were used. The study resulted in the implementation of proprietary selection algorithms for the classification of EEG records giving higher successful results, and from the proprietary selection methods discussed in this study, the proprietary selection (Ecfs) and the infinite proprietary selection (Inffs) methods given the best results among the filter-based approaches.

Anahtar Kelimeler: