融入知识图谱和课程学习的对话情绪识别  

Conversational Emotion Recognition Incorporating Knowledge Graph and Curriculum Learning

作　　者：杜金明 孙媛媛[1] 林鸿飞[1] 杨亮[1] Du Jinming;Sun Yuanyuan;Lin Hongfei;Yang Liang(College of Computer Science and Technology,Dalian University of Technology,Dalian,Liaoning 116024)

机构地区：[1]大连理工大学计算机科学与技术学院,辽宁大连116024

出　　处：《计算机研究与发展》2024年第5期1299-1309,共11页Journal of Computer Research and Development

基　　金：国家重点研发计划项目(2022YFC3301801);中央高校基本科研业务费专项资金(DUT22ZD205)。

摘　　要：对话领域情绪识别是基于对话的情感分类任务,对话数据具有口语化、主题跨度大和标签具有语义相似性的特点.口语化表现为对话中存在隐含常识和语法知识的二义性词语和省略句,导致模型难以准确建模语义信息;主题跨度大表现为不同对话场景下的文本信息丰富度差异大、情绪转移频率差异大,导致模型性能下降.提出CK-ERC模型缓解上述问题,在预训练阶段,抽取结构化数据为模型融入常识和语法知识图谱,帮助模型建模口语化信息;在微调阶段引入监督对比学习任务帮助模型识别相似情绪标签;在训练策略上设计了基于动态阈值的课程学习策略,按照文本丰富度从高到低、情绪转移频率从低到高的策略优化模型.CK-ERC模型在双人对话、多人对话、模拟对话、日常对话等多种对话模式下显著优于其他模型,在MELD和EmoryNLP数据集上获得最佳表现.Conversational emotion recognition is the task of classifying emotions based on conversations.The conversation data are characterized by colloquial language and a wide range of topics,with semantic similarities among labels.Colloquial language exhibits issues such as word ambiguity and the omission of semantic information,emphasizing the importance of common sense and grammatical knowledge in conversational emotion recognition tasks,and these factors enable the model to accurately capture semantic information.Moreover,the current challenge lies in the variations in text richness and the frequency of emotion transfer across different dialogue scenarios,which result in suboptimal classification performance.We propose CK-ERC model to address these challenges.In the pretraining phase,CK-ERC model extracts structured data to incorporate common sense knowledge graphs and grammatical knowledge graphs,aiding the model in accurately capturing colloquial information.In the fine-tuning phase,a supervised contrast learning task is introduced to help the model identify similar emotional labels.Furthermore,a dynamic threshold-based curriculum learning strategy is designed for training and optimizing the model based on text richness(from high to low)and emotion transfer frequency(from low to high).CK-ERC model demonstrates superior performance in various conversation modes,including two-person conversation,multi-person conversation,simulated conversation,and daily conversation.Particularly,CK-ERC model achieves the best performance on MELD and EmoryNLP datasets.

关 键 词：对话情绪识别 对比学习 知识图谱 课程学习 迁移学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]