NLP Based Text Analytics and Visualization of Political Speeches
Paritosh D. Katre
Paritosh D. Katre, Computer Engineering Department, Vishwakarma Institute of Information Technology, Pune, India.
Manuscript received on 14 August 2019. | Revised Manuscript received on 19 August 2019. | Manuscript published on 30 September 2019. | PP: 8574-8579 | Volume-8 Issue-3 September 2019 | Retrieval Number: C6503098319/19©BEIESP | DOI: 10.35940/ijrte.C6503.098319
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper presents a case study of implementing computational methods like Natural Language Processing (NLP) to perform Text Analytics and Visualization on political speech transcripts. The speech transcripts are published on websites, social media, and documents in large volumes and multiple languages. These transcripts are available in unstructured textual format and thus they are a part of big-data requiring analytics to derive insights from it. In this experiment, a significantly large volume of speech transcripts are analyzed and graphical visualizations are generated such as Lexical Dispersion Plot, Time Series Plot, WordCloud, Bar-Graphs using various Python libraries. The study has been useful in identifying issues highlighted across a large number of speech transcripts. So far, the linguists have tried to perform analysis using manual linguistic approaches which are extremely time-consuming and complex to understand the Political Discourse. Our experiment of applying NLP based text analytics proves to be a very efficient technique for Political Discourse Analysis (PDA).
Keywords: Natural Language Processing (NLP), Big Data, Political Discourse Analysis (PDA), Text Analytics and Visualization
Scope of the Article: Big Data