Dual Aspect Sentiment Analysis for Opinion Mining

Social media has become a widely used platform for sharing information as well as expressing feelings and emotions. The work demonstrates the method of analyzing public opinion available on social media platforms using Natural Language Processing. The proposed method extracts the major topics discussed in tweets (of data set taken), via Latent Dirichlet Allocation (LDA) topic model and identifies the underlying emotions contained in tweet using the techniques of Sentiment Analysis. The paper uses the corpus of Twitter data on admission process in India to capture the opinions of stakeholders taking part in the process. Tweeter data of Indian Institute of Technology (IIT) admission has been used to conduct the experiment. Major topics discussed in tweets and the fundamental emotions contained are obtained as results.