Data augmentation for natural language text datasets

Abstract

In this thesis, I explored the idea of using data augmentation for text datasets through developing a python application. Chapter one explains the problem statement and what I planned to achieve. Chapter two provides background information on definitions and tools used during the study. Chapter three focuses on the structure and contents of the python program as well as the features it provides. Chapter four is about the result of using the application on three datasets. Chapter five presents the summary and conclusion to the thesis, as well as some suggestions and improvements for future.