Twitter Mining in the Oil Business: A Sentiment Analysis Approach

Twitter has become a very popular communication tool among Internet users, allowing 500 millions of users to share opinions in 140 characters on different aspects of their life every day. Because of this, Twitter is a rich source of data for opinion mining and sentiment analysis that organisations can use to improve their interaction with stakeholders. In this paper, we analyse data collected from Twitter and investigate the variance that arises from using an automated sentiment analysis tool versus human classification. Our interest particularly, lies in understanding how users' motivation to post messages affects the quality of classification. The data set utilises Tweets originating from two of the world's leading oil companies, BP America and Saudi Aramco, and other users that follow and mention them, representing the West and Middle East respectively. Our results show that the two methods yield significantly different positive, natural and negative classifications depending on culture and the relationship of the poster to the two companies, calling into question the reliability of automated sentiment analysis tools for certain classes of users.