Paper Title
A Transfer Learning Approach for Multi Label Emotion Classification and Intensity

Abstract
With the increasing popularity of social media, Twitter has become one of the most popular social media websites. Twitter has hundreds of millions of active users. People can post freely on this forum and share their thoughts and opinions. For this reason, Twitter has attracted attention for analytical and research studies. Identifying and analyzing the emotions conveyed on social media content can benefit numerous sectors such as social welfare, public health, commerce, and so on. Therefore, the purpose of our research is to study the effect of posts by social media users on human emotions. Previous studies on emotion and sentiment analysis have concentrated majorly only on single-label categorization and overlooked the coexistence of more than one emotion labels in one case which we will explore. We will compare traditional methods of machine learning on emotion classification and intensity regression and our approach of transfer learning and depict how transfer learning provides near state of art results.