Comparison of various ML and DL Models for Emotion Recognition using Twitter
Hemant Kumar, Y P Gowramma, S H Manjula, D Anil, N Smitha
Abstract
Despite recent success of Sentiment Analysis in many fields to detect the emotions expressed in various social media posts are lagging. Previous studies on twitter hardly focused on lexicons and classifiers with bag-of-words model. Emotion detection with variations in expression and perception are key challenges. To address this issue, emotion classes have been defined to train the model such as fear, anger etc. Further classification approach is proposed to automatically classify the tweets to its emotion class. Our approach involves two main tasks: using machine learning classification models and neural network approach for better performance. The first task includes various classifiers like Naive Bayes, Support Vector Machine (SVM), Logistic Regression, and Random Forest are used to classify the emotions in tweets. For the second task, Neural Network Models like Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN) are used to classify the emotions in tweets. According to experimental results Support Vector Machine (SVM) outperforms compared to other models.