A Survey of Application of Artificial Intelligence Method in Text Emotion Analysis
Xiang Yang Lou · 2021
Human emotion analysis is a very extensive research field, which can be realized by a variety of applications. To engage in any project related to emotion detection, you must collect available models, data sets, and performance statistics related to it. The purpose of this review analysis is to integrate a unique and advanced set of methods proposed and/or implemented by everyone in the field. The research work of this paper mainly focuses on two aspects of classification, one is the classification based on research objects, the other is the classification based on research methods. The research object is mainly the difference between short text and normal text. The research methods mainly focus on machine learning, such as deep belief networks (DBN) and recurrent neural network (RNN). It can be seen from the comparison results that depth learning training is the best at a specific number of layers. If the number of layers is too many or too few, it will lead to insufficient accuracy or a waste of training time. Finally, this paper also puts forward feasible future research directions to provide readers with reference.