Analysis of Historical Speech Based on the Decision-Making Process of Political Leaders Using Natural Language Processing
Abstract
Historical speeches of political leaders have served as among the most powerful parts in influencing the globe. Speeches of the textual form were imprinted into history. These kinds of conversations have a significant impact on the general population and their activities in the days ahead. Additionally, if allowed unsupervised, political persons or groups may generate severe issues. In numerous situations, there may be an indicator that the governing body has to adjust its regulations and simultaneously attend to the populace. Recognizing the sentimental analysis of a political statement is vital, as they can be preliminary signs or alerts for future worldwide emergencies, alliances, battles, and future wars. This study emphasized the prime ministers of the United States and India, categorizing their speeches according to their emotional content and context. In this study, optimism is a supplementary emotion, while happiness and upset are the main ones. Beyond categorizing the speeches depending on context and emotion, among the primary tasks of this study present a database of political speeches containing 2010 speeches annotated with context-based and emotion-based historical speeches. The speeches that are being focused on are quite lengthy. This study introduces EMPOLITICIAN-Context, an adaptive voting classification technique for context-based categorization, and EMPOLITICIAN-Emotion, an adaptive voting algorithm for emotion categorization of political speeches. The suggested EMPOLITICIAN-Context approach has obtained 73.2% accuracy in the area of context-based categorizing, and the EMPOLITICIAN-Emotion approach accomplished 53.1% accuracy in categorizing the emotion of the historical speeches of political leaders.