Emotion Detection from Text Using Skip-thought Vectors

Abstract

Emotion detection from natural language has become a popular task because of the primary role of emotions in human-machine interaction. It has a wide variety of applications ranging from developing emotional chatbots to better understanding people and their lives. The problem of finding emotion from text has been handled by using lexical approaches and machine learning techniques. In recent years neural network models have become increasingly popular for text classification. Especially, the emergence of word embeddings within deep learning architectures has recently drawn a high level of attention amongst researchers. In this research, we apply a recently proposed deep learning model named skip-thought, an approach to learning fixed length representations of sentences, to face the problem of emotion detection from text. We propose a new framework that takes advantage of the pre-trained model and pre-trained word vectors. We found that skip-thought vectors are well suited for emotion detection task. The results of the performance evaluation are encouraging and comparable to related research.