Comparative analysis of machine learning algorithms for Classification of US Airlines Tweets

Sentiment classification (SC) task is well favoured in e-commerce field as the reviews posted by users, classified into positive and negative categories, help the business holders to understand their current market and other's market strategy and settings as well. This task of SC is widely encouraged by various researchers for different domains such as movie reviews, political reviews, product reviews, airlines and so on. This paper aims to classify the tweets that are shared by the American flight travellers into positive and negative class so that the other people can utilize this information to take their journey related decisions. The task has been accomplished using three machine learning classifiers to classify the tweets data that is collected from Kaggle site. It is seen from the experimental results that SVM outperforms the LR and MNB baselines by 0.9% and 6.23% accuracy respectively.