Syntactic pattern recognition applied to sleep EEG staging (original) (raw)

Pattern Recognition Letters, 1989

Abstract

Abstract Syntactic pattern recognition approach was applied to classification of EEG segments belonging to sleep stage I and stage REM. Feature extraction was carried out using adaptive segmentation of the EEG signal, and the process of primitive recognition was implemented through unsupervised fuzzy clustering. Grammatical inference was done automatically constructing canonical derivative finite-state grammers for the two training sets. For the classification, a minimum-distance error-correcting parser was used, to account for substitution errors in the process of primitive recognition.

isak gath hasn't uploaded this paper.

Let isak know you want this paper to be uploaded.

Ask for this paper to be uploaded.