Convolutional Neural Network Combined with Emotional Dictionary Apply in Chinese Text Emotional Classification

Abstract

AbstractAt present, convolutional neural network has achieved good results in text emotional classification, but in several common models, it does not make use of a large number of prior knowledge that human society has now acquired. This paper proposes a new CNN model based on emotional dictionary: emotional knowledge-CNN (EK-CNN), which uses emotional dictionary as additional knowledge to improve the performance of the model in emotional classification. The model has been validated on three real data sets, and is superior to the existing technology model for text emotion classification.KeywordsNatural Language Processing (NLP)Emotional dictionaryKnowledgeConvolutional Neural Network (CNN)