Sentiment Analysis using Ensemble Technique on Textual and Emoticon Data

Nowadays social media is very much popular as people can convey their feelings or recapitulate. Many people express through emotions. Sentiment Analysis using text and emoticons has become leading and challenging research topic in recent world. In the past, text, image, and emoticons were classified; however, there are few research on text and emoticon in combination. In this work, the classification of sentiments was done on both text and emoticons data. The data were on airlines feedback and collected from UCI repository. The data were classified as positive, negative and neutral class using machine learning (ML), deep learning (DL) and ensemble classifiers based on three types of features such as TF-IDF, Bag-of-Words, and N-grams. The comparative evaluation of results shows machine learning to be outperformer over DL and ensemble classifiers.