Ascertaining Public Opinion Through Sentiment Analysis

Over the years, the evolution of the internet has played an important role in interconnectivity amongst people from every walk of life. One of the consequences to arise from this evolution is the emergence of online web forums and social media outlets. These forums act as a resource full of rich text-based data ready to be mined. Sentiment analysis is a concept that deals with people's opinion, emotions, attitudes towards any particular field or topic. As of 2019 it is one of the most discussed and researched area in the NLP (Natural Language Processing) and is also studied (widely) in data/web/text-based mining. Sentiment analysis at its current stage is widespread among many other fields (let alone computer science) including management and social media marketing due to its importance to business and society. This paper focuses on the workings of a machine learning based model that monitors public opinion on trending topics on the social media platform, Twitter, to arrive at the emotional aspect of the opinion that is being conveyed. This is in turn achieved by the NLP approach, that mines the data for emotional cues based on predefined keywords and ascertains the polarity of the public view on the topic.