Human annotation and interpretation of public sentiments about JIO coin marked in social networks using machine learning algorithms

Abstract

The basic errand in Sentiment Analysis is to categorize the orientation of a given review and subsequently identifying whether the sentiment implied is positive, negative or fair. In this article the authors present the following lines of experimentation and outcomes. One is related to human annotation of Tweets and assessment of their quality and dataset properties. Another is about training sentiment classifiers, their performance and comparisons. The authors' presents a comprehensive assessment about various supervised machine learning techniques to interpret the public sentiments about ‘Jio Coin’ marked in social networks. Various evaluation measures like Precision, Recall, F-Score, Matthews Correlation Coefficient, Jaccard Index and Kappa statistics depicts the efficiency rate of the models in the datasets. The learning time and the predicting time taken by various classifiers depicted in the article helps to choose the classifier that suits well if time is a constraint.