Sentiment Analysis of Garments Workers Using Machine Learning

Abstract

Sentiment Analysis (SA) is a very popular field for researchers in the field of text mining. Computational treatment of opinions, sentiments and subjectivity of texts are shown by sentiment analysis. In this study I examined the sentiment of the most overlooked group of people known as the Garments Workers. First I present an unsophisticated dataset that uses novel reviews of workers to capture their satisfaction. I went to some companies and took data of the Garments Workers manually. Then, I created a document by aggregating worker emotions of the companies and measuring employee sentiment as to whether they are happy or unhappy using the Ensemble Learning approach to count happy and unhappy terms. Finally, I defined worker satisfaction by using a 10 fold validation method. The results of my model suggest that it may be beneficial for investors to incorporate a measure of worker satisfaction into their method for forecasting earnings.