Sleep stage classification using fuzzy long short-term memory

This study investigates the performance of Fuzzy Long Short-Term Memory (FLSTM) for sleep stage classification. The proposed FLSTM consists of the Fuzzy C-Means Clustering which functions as feature representation, and Long Short-Term Memory (LSTM) as the final classifier. The performance was evaluated based on accuracy, precision, and F-measure by testing some cluster number values from Fuzzy C-Means Clustering. The output of this clustering becomes input for Long Short-Term Memory. The result shows that the best performance achieved when using as much as 9 clusters.