Twitter Sentiment Analysis on Automotive Companies

Abstract

Many users would use social media to express their opinions on their products or services. The expression can be good or bad. This project proposed a sentiment analysis on the automotive company where the users' opinions are analysed through this feedback. The data are collected from the social media of Twitter and followed by data mining techniques which are tokenization, removing stop words, and stemming. A sentiment classifier is implemented after the data have been converted into valuable data. Naïve Bayes classification is employed in this project by using Python language. Based on the dataset that we use, the article may analyse market demand for the automobile industry. According to the findings, Honda and Mazda had the highest positive sentiment, with more than 85 percent. This project is beneficial to the automotive industry, especially to teams' production. This finding supports a better understanding between the industry and their customer, to enhance the business strategies and find out the weaknesses.