Aspect-Based Twitter Sentiment Classification
Hsiang Hui Lek, Danny Poo · 2013
Due to the popularity of Twitter, sentiment classification for Twitter has become a hot research topic. Previous studies have approached the problem as a tweet-level classification task where each tweet is classified as positive, negative or neutral. However, getting an overall sentiment might not be useful to organizations which are using twitter for monitoring consumer opinion of their products/services. Instead, it is more useful to determine specifically which aspects of the products/services the users are happy or unhappy about. This paper proposes an aspect-based sentiment classification approach to analyze sentiments for tweets. To the best of our knowledge, we are the first to perform sentiment analysis for Twitter in this manner. We conducted several experiments and show that by incorporating results from the aspect-based sentiment classifier, we are able to improve existing tweet-level classifiers. The experimental results also demonstrated that our approach outperforms existing state-of-the-art approaches.