Multilabel Text Classification Using Transformers

Social media platforms are widely popular for communicating ideas, comments, and feelings. In a new digital age of communication, it is becoming more important to characterize online comments into their respective emotions to follow their intended meaning without visual cues that would normally be applied during in-person communication. Multilabel text classification is a method used to identify the multiple attitudes and emotions of a single comment. Specifically, multilabel classification can categorize a comment into numerous emotion classes whereas traditional classification would assign a comment the single best fit emotion. Using artificial intelligence and machine learning techniques, text can be classified and labeled with multiple emotions.