Multilingual Lexicon based Approach for Real-Time Sentiment Analysis
Swati Sharma1, Mamta Bansal2
1Swati Sharma**, PhD Scholar, Shobhit University, AP at MIET, Meerut, India.
2Mamta Bansal , Professor at Shobhit University, Meerut, India.
Manuscript received on May 25, 2020. | Revised Manuscript received on June 29, 2020. | Manuscript published on July 30, 2020. | PP: 984-989 | Volume-9 Issue-2, July 2020. | Retrieval Number: B3997079220/2020©BEIESP | DOI: 10.35940/ijrte.B3997.079220
Open Access | Ethics and Policies | Cite | Mendeley
© 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: The information on WWW has mounted to a greater height, overriding to fledgling analysis in the direction of sentiments using Artificial Intelligence. Sentiment Analysis deals with the calculus exploration of sentiments, opinions and subjectivity. In this paper, multilingual tweets are analyzed for identifying the polarities of various political parties like AAP, BJP, Samajwadi, BSP and Congress; so that the users will get an idea that to which party they should give their vote. The data is being analyzed using Natural Language Processing. Using different smoothening techniques, noise is removed from data, classified by using Machine learning algorithms and then the accuracy of the system is gauged using various evaluation precision measures. The central premise of this research is to benignant common people and politicians both. For common people; is for deciding their precious vote, to which party to give will be good for themselves and nation too. For politicians; they will have an idea about themselves i.e. after seeking the polarities of different parties, the politicians will have an idea which party is preferable and which is not preferable, so that the politicians can work accordingly. The system shows comparison among VADER and SVM algorithm; and SVM algorithm showed 90% accuracy.
Keywords: Lexicon, NLP, SVM, VADER