.

ISSN 2063-5346
For urgent queries please contact : +918130348310

ENHANCING THE ACCURACY IN CLASSIFYING HUMAN EMOTION VIA SPEECH RECOGNITION USING NOVEL MULTILAYER PERCEPTRON COMPARED WITH LOGISTIC REGRESSION

Main Article Content

S.Devakar, D. Beulah David
» doi: 10.31838/ecb/2023.12.sa1.393

Abstract

Aim: To enhance accuracy of human emotion classification via speech recognition using Novel Multilayer Perceptron compared with Logistic Regression. Materials and Methods: This study contains 2 groups i.e Novel Multilayer Perceptron (MLP) and Logistic Regression (LR). Each group consists of a sample size of 6. Gpower software is used to determine sample size with pretest power value 0.8 and alpha is 0.05 with the p = 0.138. Result: The Novel Multilayer Perceptron 70.37% more accurate than Logistic Regression 60.99% in classifying human emotion. Conclusion: The Novel Multilayer Perceptron is significantly better than Logistic Regression in classifying human emotion via speech recognition.

Article Details