Examining the emotional tone in politically polarized Speeches in India: An In-Depth analysis of two contrasting perspectives

Abstract

Sentiment analysis is an area of natural language processing (NLP) which uses machine learning algorithms to interpret textual data. Recently, sentiment analysis has become an increasingly popular trend within social sciences studies led by computer scientists. NLP techniques have recently been applied to several political science topics, such as topic modeling and supervised machine learning. These techniques help political scientists recognize underlying ideology or misinformation present in texts as well as detect changes over time in political messages.  This study examines the application of topic modeling on political speeches to understand the recurring themes present in them. Recent speeches for last one year for the prime minister of India and key opposition party leader has been selected for the same. It delves into the coherence of the topics generated through the topic modeling method and the influence of text size on the coherence of political speech themes. The analysis focuses on determining the presence of positive or negative sentiments in public speeches and the utilization of Latent Dirichlet Allocation (LDA) and Coherence Measures as techniques for modeling and analyzing public speeches. The LDA algorithm operates by analyzing a speech and determining the number of words it includes, then identifying the topics. Paper, find one views to be extremely positive and other one with negative sentiments. Also, paper shares some of the key topics for the same