Fig. 5: The Receiver Operating Characteristic (ROC) curves show the implementation of different MLAs for classifying the contrastive tones (EH, H, M, L, and MR).
From: Multi-class identification of tonal contrasts in Chokri using supervised machine learning algorithms
I [a–f] represent the curve generated on the female speakers for different MLAs, and II [a–f] show the curve generated on the male speakers data for different MLAs. The different MLAs include a Decision Tree (DT), b K-Nearest Neighbors (KNN), c Logistic Regression (LR), d Naive Bayes (NB), e Random Forest (RF), and f Support Vector Machine (SVM).
