Utilizing Machine Learning Algorithms to Detect Emotions from Tweets
Christopher Reynard Julian, Gian Reinfred Athevan, Nathanael Geoffrey Hanjaya, Henry Lucky, Derwin Suhartono · 2022
Emotion detection has been a popular topic for a long time. The need to develop and improve the technologies has grown throughout the year. This is mostly because of the vast implementation in marketing, education, security, healthcare, and other fields. Emotion can be detected variously, from gestures, text, speech, facial expressions, and others. Today, the emotion detection technology is still not perfect, even with the huge access to textual and visual data. This is mostly because of the complexity of human emotions that can’t be easily captured by a machine. Therefore, a reliable method is required in order to create an accurate emotion detection system. A few of the popular machine learning methods that are implemented for emotion detection are Support Vector Machine, Naïve Bayes, and Logistic Regression. From our experiment, SVM is considered to be the best methods that have demonstrated high accuracy, especially on a quality data set.