Analysis of Indian News with Corona Headlines Classification

With the advent of the world wide web, the world has seen an explosion in the amount of information that is available online. People stay informed about the national and international affairs through online news which is readily available and portable allowing ease of access. These news pieces tend to shape people’s thoughts and provoke emotions, which may be positive, neutral or negative, without them realizing their effect. The objective of this work is to create a hybrid model that can analyze the overall effect of digital news content in India. The hybrid approach of sentiment analysis encompasses lexicon and machine learning algorithms as well as a self-created scored corpus of corona related words to classify all sorts of headlines. The labelled dataset is used to train decision tree and random forest algorithms. They are evaluated based on their accuracy scores, classification reports and confusion matrices. The results prove that both the algorithms perform well on the dataset and that the Indian media highlighted neutral news the most. This finding can be very useful for the Indian news agencies since they can alter their reporting strategies to create an impact of their choice on the readers’ minds and thus, increase the readership.