Text-based Emotion Detection: A Review

Abstract

Textual language is the most natural carrier to express the emotions of human beings. Emotion Recognition and analysis is a major challenging and emerging topic in the research area of Natural Language Processing (NLP) due to its significant academic and commercial potential. Along with the growth of the Internet, people use social media or digital platforms to share their feelings and feedback. Hence, it is important for a machine to understand the emotions, feedback, and textual dialogues to provide emotionally aware responses to users in today’s digital world. This survey has been inspired on the well-known fact that, despite there is a lot of work on emotional detection systems, a lot of work is expected to be done yet. The increment of these systems is due to the large amount of emotional data available in Social Web. Detecting emotions from text have attracted the attention of many researchers in computational linguistics because it has a wide range of applications. This paper presents a literature review of existing literature published between 2013 and 2023. This review has meticulously examined the various emotion models, techniques, datasets, research and evaluation metrics in the detection of emotions from the text useful for researchers in carrying out emotion detection activities. We also discussed about the different challenges with their future direction.