[1]郑思露,程春玲,毛 毅.融合实体信息和时序特征的意图识别模型[J].计算机技术与发展,2022,32(11):171-176.[doi:10. 3969 / j. issn. 1673-629X. 2022. 11. 025]
　ZHENG Si-lu,CHENG Chun-ling,MAO Yi.An Intention Recognition Model Combining Entity Information and Temporal Features[J].,2022,32(11):171-176.[doi:10. 3969 / j. issn. 1673-629X. 2022. 11. 025]
点击复制

融合实体信息和时序特征的意图识别模型()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年11期
页码:
171-176
栏目:
人工智能
出版日期:
2022-11-10

文章信息/Info

Title:
An Intention Recognition Model Combining Entity Information and Temporal Features
文章编号:
1673-629X(2022)11-0171-06
作者:
郑思露; 程春玲; 毛 毅
南京邮电大学 计算机学院,江苏 南京 210023
Author(s):
ZHENG Si-lu; CHENG Chun-ling; MAO Yi
School of Computer Science,Nanjing University of Posts & Telecommunications,Nanjing 210023,China
关键词:
深度学习; 意图识别; 特征融合; 实体信息; 时序特征; gMLP
Keywords:
deep learning; intention recognition; feature fusion; entity information; time series characteristics; gMLP
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 11. 025
摘要:
人机对话意图识别旨在通过人机之间简短的对话识别出用户意图,通过对话文本的分类进而实现意图的识别。针对人机对话中因篇幅短导致语境匮乏和因对话随意性导致意图模糊的问题,提出了一种融合实体信息和时序特征的人机对话意图识别模型。 在文本表示阶段,通过捕捉对话中实体信息来增强文本语义表达,并利用双向注意力机制动态生成符合语境的文本表示;并利用双向 GRU 提取对话上下文的时序特征来获取上下文意图之间的关系;通过级联多层gMLP,利用其内部空间控制单元自适应融合实体信息和时序特征,从而提升意图识别的准确率。 为验证所提模型在多种任务上的效果,在不同意图识别任务数据集 CCKS2018 和 SMP2018 上进行实验,分别取得了 90. 6% 和 93. 7% 的准确率,对比 CLSTM、DBN、Attention-RNN 等具有代表性的模型,均有 3% 以上性能的提升。
Abstract:
The purpose of human - machine dialogue intention recognition is to identify the user ’s intention through the brief dialoguebetween human and machine, and then realize the intention recognition through the classification of the dialogue text. Aiming at theproblems of lack of context due to short length and ambiguity of intention due to randomness of dialogue in man -machine dialogue,aman-machine dialogue intention recognition model integrating entity information and temporal features is proposed. In the stage of textrepresentation,the text semantic expression is enhanced by capturing the entity information in the conversation, and the bidirectionalattention mechanism is used to dynamically generate the text representation in accordance with the context. Bidirectional GRU is used toextract the temporal features of the dialogue context to obtain the relation between context intentions. In order to improve the accuracy ofintention recognition,multi-layer gMLP is used to adaptively fuse entity information and temporal features with its internal spatial controlunit. In order to verify the effectiveness of the proposed model on multiple tasks, experiments were carried out on CCKS2018 andSMP2018 data sets of different intention recognition tasks, and the accuracy was 90. 6% and 93. 7% , respectively. Compared withCLSTM、DBN、Attention-RNN and other representative models,the performance was improved by more than 3% .

相似文献/References:

[1]陈强锐,谢世朋.基于深度学习的肺部肿瘤检测方法[J].计算机技术与发展,2018,28(04):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
　CHEN Qiang-rui,XIE Shi-peng.Lung Cancer Detection Method Based on Deep Learning[J].,2018,28(11):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
[2]施泽浩,赵启军.基于全卷积网络的目标检测算法[J].计算机技术与发展,2018,28(05):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
　SHI Ze-hao,ZHAO Qi-jun.Object Detection Algorithm Based on Fully Convolutional Neural Network[J].,2018,28(11):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
[3]黄法秀,张世杰,吴志红,等.数据增广下的人脸识别研究[J].计算机技术与发展,2020,30(03):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
　HUANG Fa-xiu,ZHANG Shi-jie,WU Zhi-hong,et al.Research on Face Recognition Based on Data Augmentation[J].,2020,30(11):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
[4]陈浩翔,蔡建明,刘铿然,等. 手写数字深度特征学习与识别[J].计算机技术与发展,2016,26(07):19.
　CHEN Hao-xiang,CAI Jian-ming,LIU Keng-ran,et al. Deep Learning and Recognition of Handwritten Numeral Features[J].,2016,26(11):19.
[5]高翔,陈志,岳文静,等.基于视频场景深度学习的人物语义识别模型[J].计算机技术与发展,2018,28(06):53.[doi:10.3969/ j. issn.1673-629X.2018.06.012]
　GAO Xiang,CHEN Zhi,YUE Wen-jing,et al.Human Semantic Recognition Model Based on Video Scene Deep Learning[J].,2018,28(11):53.[doi:10.3969/ j. issn.1673-629X.2018.06.012]
[6]贺飞翔,赵启军. 基于深度学习的头部姿态估计[J].计算机技术与发展,2016,26(11):1.
　HE Fei-xiang,ZHAO Qi-jun. Head Pose Estimation Based on Deep Learning[J].,2016,26(11):1.
[7]徐 融,邱晓晖.一种改进的 YOLO V3 目标检测方法[J].计算机技术与发展,2020,30(07):30.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 007]
　XU Rong,QIU Xiao-hui.An Improved YOLO V3 Object Detection[J].,2020,30(11):30.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 007]
[8]曾志平[] [],萧海东[],张新鹏[]. 基于DBN的金融时序数据建模与决策[J].计算机技术与发展,2017,27(04):1.
　ZENG Zhi-ping[] [],XIAO Hai-dong[],ZHANG Xin-peng[]. Modeling and Decision-making of Financial Time Series Data with DBN[J].,2017,27(11):1.
[9]李全兵,文 钊*,田艳梅*,等.基于 WGAN 的音频关键词识别研究[J].计算机技术与发展,2021,31(08):26.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 005]
　LI Quan-bing,WEN Zhao *,TIAN Yan-mei *,et al.Research on Audio Keywords Recognition Based on WassersteinGenerative Adversarial Network[J].,2021,31(11):26.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 005]
[10]李宏林. 分析式纹理合成技术及其在深度学习的应用[J].计算机技术与发展,2017,27(11):7.
　LI Hong-lin. Analyzed Texture-synthesis Techniques and Their Applications in Deep Learning[J].,2017,27(11):7.

更新日期/Last Update: 2022-11-10