Abstract
The study of sentiment analysis aims to explore the occurrences of opinion words in a document, and their orientation and strength. Most existing researches focus on writer's emotions, which are the feelings that authors were expressing. In contrast, news articles are usually described in objective terms, in which opinion words are rare. In this paper, we propose to discover the reader's feelings when they read news articles. First, different feature units are extracted such as unigrams, bigrams, and segmented words. We compare several feature selection methods including document frequency variation, information gain, mutual information, and chi-square test, to select the candidate features for sentiment classification. Then, the performance for multi-class classification and multiple binary classification are evaluated. The experimental results on real news articles show the effectiveness of the proposed method for discovering reader's emotions triggered from news articles. Further investigation is needed to validate the performance in larger scales.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
