Positioning of Trending Topics and Analyzing the Tweets in Social Network using Deep Learning
H. Aswini1, J. Jayabharathy2, G. Balamurugan3

1H. Aswini*, is currently pursuing master degree program in Computer Science and Engineering in Pondicherry Engineering College, India.
2J. Jayabharathy, is currently working as Associate Professor in Computer Science and Engineering in Pondicherry Engineering College, India.
3G. Balamurugan, is currently pursuing research in Computer Science and Engineering in Pondicherry Engineering College, India.

Manuscript received on June 08, 2020. | Revised Manuscript received on June 25, 2020. | Manuscript published on June 30, 2020. | PP: 1308-1312 | Volume-9 Issue-5, June 2020. | Retrieval Number: E9876069520/2020©BEIESP | DOI: 10.35940/ijeat.E9876.069520
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: In the Digital time, Twitter has developed to turn into a significant web based life to get to quick data about unique themes that are slanting in the public eye. In later, identification of topical substance utilizing classifiers on Twitter can sum up well past the enormous volume of prepared information. Since access to Twitter information is holed up behind a restricted pursuit API, normal clients can’t have any significant bearing these classifiers legitimately to the Twitter unfiltered information streams. Or maybe, applications must pick what substance to recuperate through the pursuit API before sifting that content with topical classifiers. In this manner, other than these lines, it is basic to scrutinize the Twitter API near with the proposed topical classifier in a manner that limits the measure of adversely arranged information recovered. In this paper, we propose a succession of inquiry enhancement strategies utilizing Machine learning with the assistance of CNN that sum up thoughts of the most extreme inclusion issue to discover the subclass of question articulations inside as far as possible. It is utilized to cover most of the topically pertinent tweets without relinquishing accuracy. Among numerous bits of knowledge, proposed techniques fundamentally outflank the scientific classification dependent on the tweets and arrange the best of the tweets and pessimistic tweets in Twitter.
Keywords: Cynical tweets, Precision, Social media, Twitter API, Topical content.