Emotional Dialogue Generation Based on Conditional Variational Autoencoder and Dual Emotion Framework
Open Access
- 26 December 2020
- journal article
- research article
- Published by Wiley in Wireless Communications and Mobile Computing
- Vol. 2020, 1-10
Abstract
An excellent dialogue system needs to not only generate rich and diverse logical responses but also meet the needs of users for emotional communication. However, despite much work, these two problems have not been solved. In this paper, we propose a model based on conditional variational autoencoder and dual emotion framework (CVAE-DE) to generate emotional responses. In our model, latent variables of the conditional variational autoencoder are adopted to promote the diversity of conversation. A dual emotion framework is adopted to control the explicit emotion of the response and prevent the conversation from generating emotion drift indicating that the emotion of the response is not related to the input sentence. A multiclass emotion classifier based on the Bidirectional Encoder Representations from Transformers (BERT) model is employed to obtain emotion labels, which promotes the accuracy of emotion recognition and emotion expression. A large number of experiments show that our model not only generates rich and diverse responses but also is emotionally coherent and controllable.Keywords
Funding Information
- Guangxi Key Laboratory of Image and Graphic Intelligent Processing Project (GIIP2003, AB20238013, ZY20198016, 2019GXNSFFA245014, 6202780103, 61772149, 61762028, U1701267)
- Guangxi Science and Technology Project (GIIP2003, AB20238013, ZY20198016, 2019GXNSFFA245014, 6202780103, 61772149, 61762028, U1701267)
- National Natural Science Foundation of China (GIIP2003, AB20238013, ZY20198016, 2019GXNSFFA245014, 6202780103, 61772149, 61762028, U1701267)
This publication has 9 references indexed in Scilit:
- A worker-selection incentive mechanism for optimizing platform-centric mobile crowdsourcing systemsComputer Networks, 2020
- Dynamic Control of Fraud Information Spreading in Mobile Social NetworksIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2019
- Incentive Mechanisms for Crowdblocking Rumors in Mobile Social NetworksIEEE Transactions on Vehicular Technology, 2019
- Generating Emotional Controllable Response Based on Multi-Task and Dual Attention FrameworkIEEE Access, 2019
- Prior Knowledge-Based Probabilistic Collaborative Representation for Visual RecognitionIEEE Transactions on Cybernetics, 2018
- A Private and Efficient Mechanism for Data Uploading in Smart Cyber-Physical SystemsIEEE Transactions on Network Science and Engineering, 2018
- Emotional Human-Machine Conversation Generation Based on Long Short-Term MemoryCognitive Computation, 2017
- Online Low-Rank Representation Learning for Joint Multi-Subspace Recovery and ClusteringIEEE Transactions on Image Processing, 2017
- Long Short-Term MemoryNeural Computation, 1997