Polarity Detection of Sentiment Scoring Methodusing Dempstershafer Theory
M. Edison1, A. Aloysius2
1M. Edison, Assistant Professor, Apollo Arts and Science College, Chennai (Tamil Nadu), India.
2Dr. A. Aloysius, Assistant Professor, St. Joseph’s College Autonomous, Tiruchirappalli (Tamil Nadu), India.
Manuscript received on 21 May 2019 | Revised Manuscript received on 07 June 2019 | Manuscript Published on 15 June 2019 | PP: 262-266 | Volume-8 Issue-1S2 May 2019 | Retrieval Number: A00600581S219/2019©BEIESP
Open Access | Editorial and Publishing 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: Sentiment Analysis (SA) is a big task to measure the people opinion. The aim of the SA is to obtain the essential viewpoint of text, which could be opinions, blogs, reviews, online rating comments etc. Nowadays, most of the peoples are familiar for using internet to express their opinions. However, the opinions are classified in a different way like positive, negative and neutral and assigned score to the sentimentword like +1, -1, and 0 respectively. Nevertheless, the sentiment score have been assigned formally in the existing works. Some of the works the sentiment scores are assigned based on the threshold value 0.5.In that case, several existing works were applied this value for polarity detection.The limitation of the existing works given more attention to compute the score of a sentiment. Therefore, the new algorithm proposed namely Senti_Demp_Score, which performs to measure the sentiment score based on the DempsterShaper Theory (DST). The DST perform to calculatethe sentiments in a sentence and sum of all the sentiments with a category, then they converted into percentage. Concurrently, the percentage values are summed with a category and subtractedthe percentage score which gives the accuracy of the Senti_Demp_Score is 0.7756 like in percentage 77.56%. Mainly, the algorithm Senti_Demp_Score has given better solution to detect the polarity of the sentiment and measure the sentiment score in a right way.
Keywords: Sentiment analysis, Polarity, Dempster Shafer Theory, Sentiment Score.
Scope of the Article: Algorithmic Game Theory