Determining the Sentiments of Bi-lingual Tweets Using Convolutional Neural Network and Support Vector Machine

K. Nithya and Dr.M. Krishnamoorthi

This paper aims at identifying the opinion of Tamiland English tweets posted in social media. It classifies the tweets as positive, negative and neutral.Social media data are increasing rapidly and these contents are mostly posted in English language as well as in Regional languages.Most of the recommendation systems omit the regional language texts and designed only foranalysing English language. A Novel deep learning approach proposes Bi-lingual text opinion identification. The dataset for this work is gathered from twitters public timeline. Features related to tweets are extracted (Tweet specific features i.e Hashtags and Emoticons) and also features from Sentiwordnet are extracted. Here, in this work, Machine learning algorithm such as Support Vector Machine (SVM) and Convolutional Neural Network (CNN) is used for identifying opinion. It is also compared with various classification techniques for identifying Tamil language tweets and English language tweetsopinion.

Volume 12 | Issue 1

Pages: 80-89

DOI: 10.5373/JARDCS/V12I1/20201011