Twitter based model for emotional state classification

With the advent and the subsequent rise of social network, there has been a surge of users expressing their emotions and daily feelings leveraging the social media platform. Each unit time, such data that is generated in monumental sizes, can be utilized to accurately detect one's emotional state. Twitter tweets is seen as a great source of information that can be exploited to build highly accurate and relevant emotion classifiers [1]. Through this paper, we aim to propose a model to classify an individual's recent emotional state into eight predefined states. We also subsequently compare the results and accuracy of SVM, KNN, Decision Tree & Naive Bayes algorithm to implement and justify our prescribed approach.