The analysis of the Twitter dataset for emotion intensity classification task using various machine learning algorithms. The analysis has been done on a cut down version of the original dataset to fasten the process. Initially, Conventional Machine Learning techniques such as Logistic Regression, Random Forest, Naive Bayes, SVM (Linear & Polynomial) and Neural Networks had been used and ultimately, we picked the model that produced the best accuracy for both datasets (public & private).
Furthermore, a Deep Machine Learning Model such as a fully connected ANN Dense Model has been used and predictive analysis performed on both, the public and private datasets.