Triangulated Sentiment Analysis of Tweets

Abstract

Social media platforms like Twitter present an unprecedented opportunity for customer relationship management by analysing the ongoing discussions about business events such as a service outage. These opinions have been analysed for sentiment with lexicon-based and machine learning approaches. Both methods view sentiment as either positive, neutral, or negative. According to the psycholinguistic approach, text sentiment is more continuous reflecting more naturally how we experience emotions. We compare these three approaches with a Twitter dataset collected during a service outage. Contrary to our expectation, we find that the language used in tweets is not very negative or emotionally intense. This research therefore contributes to the sentiment analysis discussion by dissecting three methods and discussing how and why they arrive at differing results. The selected research context provides an illuminating case about service failure and recovery.