Dialogue emotion detection method and device, storage medium and electronic equipment

Abstract

The invention belongs to the technical field of natural language processing, and relates to a dialogue emotion detection method and device, a storage medium and electronic equipment. The method comprises the following steps: obtaining a previous capsule vector and a current capsule vector respectively corresponding to a previous round of dialogue text and a current dialogue text by utilizing a gated recurrent neural network model; normalizing the previous capsule vector and the current capsule vector through a normalization exponential function to obtain an interactive capsule vector corresponding to the current capsule vector; and performing nonlinear compression transformation on the interactive capsule vector by utilizing a dynamic routing algorithm to obtain an emotion capsule vector so as to determine the emotion of the current dialogue text. According to the method, the emotion of the current dialogue text is determined through interactive processing of the last round of dialoguetext and the current dialogue text, the context relationship and the utterance importance relationship of the current dialogue text can be accurately captured, influence factors of emotion detectionare considered in multiple dimensions, the accuracy of emotion detection is improved, and the interpretability of emotion detection is also improved.

Description

technical field [0001] The present disclosure relates to the technical field of natural language processing, and in particular to a dialogue emotion detection method, a dialogue emotion detection device, a computer-readable storage medium, and electronic equipment. Background technique [0002] Dialogue emotion recognition is of undeniable importance in affective dialogue systems, where bots are able to understand users' moods and emotions to generate emotionally coherent and empathetic responses. At present, many scholars have done related work in the dialogue emotion recognition task. [0003] However, in the process of implementing the present invention, the inventors found that most of the existing research work uses recurrent neural networks to process dialogue sequences, but these works ignore the interaction between multiple rounds of dialogue, resulting in low accuracy of dialogue emotion detection and poorly explainable questions, which play a crucial role in the u...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present disclosure is to provide a dialog emotion detection method, a dialog emotion detection device, a computer-readable storage medium, and an electronic device, thereby at least to a certain extent overcoming the low accuracy of emotion detection and the possibility of Poor explanatory issues

Images