MS Thesis
Abstract:
This work is an attempt to study polarization on social media data, in the Indian context. We focus on multiple controversial and talked about events and topics in the context of India, namely 1) the Sabarimala Temple (located in Kerala, India) incident which became a nationwide controversy when two women under the age of 50 secretly entered the temple breaking a long standing temple rule that disallowed women of menstruating age (10-50) to enter the temple and 2) the Indian government's move to demonetise all existing 500 and 1000 denomination banknotes, comprising of 86% of the currency in circulation, in November 2016 3) The Pulwama-Balakot incidents and its effect on popular sentiment towards India's Prime Minister Narendra Modi 4) Social media activity on events surrounding multiple lynching incidents from 2017 to 2019 and 5) the Indian Economy from 2013 to 2019 and popular response towards it on social media. We gather tweets from news media, and from common people on Twitter around these events in various time periods. Then, we pre-process and annotate them with their sentiment polarity and emotional category, and analyse trends to help us understand changing polarity over time around controversial events. While this study tries to find trends and patterns to decode the aspect of polarization in India's social fabric, it's also a case study on all these events and how India reacted to each one of them. The tweets collected are in English, Hindi and code-mixed Hindi-English.
Key Contributions:
- We pick up five important events and find patterns and trends in changing opinions, before, during and after the event.
- We collect more than 2 million tweets, annotate them with a sentiment analyzer and an emotional analysis lexicon to conduct the experiments.
- We show how a controversial event leads to a significant increase in polarization on social media even months after the event.
- We also analyze how Hindi and English news media in India may have different reactions to an event, possibly due to the difference in audience.
- We also study the effectiveness of sentiment analysis using tools built for English language on data translated from Hindi to English and code-mixed data transliterated and translated to English.