Masters Theses

AN EXPERIMENTAL COMPARISON OF TWO MACHINE LEARNING APPROACHES TO INVESTIGATE GENDER DIFFERENCES IN EMOTION CLASSIFICATION

Keywords and Phrases

Business Analytics; Data Science

Abstract

"Correctly identifying an emotion has always been challenging for humans, not to mention machines! In this research, I used machine learning to classify human emotion. Emotional differences between genders are well documented in fields like psychology. I hypothesize that gender will also influence accuracy of classifying emotion with machine learning. To recognize emotions on images, I used Python, OpenCV, and the Cohn-Kanade dataset, which is a face dataset containing emotions. OpenCV (Open Source Computer Vision) is a library of programming functions. OpenCV has some functions of face recognition, one of which is using a technique called Fisher Face to get an eigenfaces, which are the eigenvectors of the set of faces. Two different machine learning approaches were tested in an experimental study. In one approach, emotions from both genders were used to train the machine. In another approach, the two genders were separated, and two separate machines were used to learn the emotions of the two genders. The results show that the approach where the genders were separated produces a higher accuracy in classifying emotions. I also observe that training sample sizes have different impact on the two approaches"--Abstract, p. iii

Advisor(s)

Siau, Keng, 1964-

Committee Member(s)

Nah, Fiona Fui-Hoon, 1966-
Chen, Langtao

Department(s)

Business and Information Technology

Degree Name

M.S. in Information Science and Technology

Publisher

Missouri University of Science and Technology

Publication Date

Fall 2017

Pagination

ix, 44 pages

Note about bibliography

Includes bibliographical references (pages 39-43)

Rights

© 2017 Wangchuchu Zhao, All Rights Reserved

Document Type

Thesis - Open Access

File Type

text

Language

English

Thesis Number

T 12174

This document is currently not available here.

Share

 
COinS