当前位置：文档库 › 面向出行领域的任务型对话系统研究

面向出行领域的任务型对话系统研究

Abstract

Nowadays,task-oriented dialogue system is currently a research hotspot in academia and industry, and has received extensive attention from domestic and foreign scholars and companies in recent years. Task-oriented dialogue system can help users complete task-based instructions through multiple rounds of dialogues, such as querying bus routing, restaurant reservations, and recommendations for tourist attractions. The current mainstream research methods including dividing dialogue system into three modules of natural language understanding, dialogue management and natural language generation, furthermore, more and more studies are focusd on using end-to-end neueral network on task-oriented dialogue system.

This paper focuses on multi-round task-oriented dialogue system for the travel domain. The main modules of the dialogue system include questions standardization module, dialogue understanding module, dialogue management module, and dialogue generation module. The main research contents of this topic include:

(1) In this paper, we establish a database of intent-detection corpus and slot-filling corpus for the travel domain, so as to facilitate the intent-detection and slot-filling experiments.

(2) This paper attempts to introduce question normalization module before the dialogue understanding module, aiming at solving the problem of garbled characters, emoji expressions, demonstrative pronouns and sentence substructures that may appear in the question sentence. We tried to combine with the above slot information for deletion and substitution, filling and other operations.After the standardization of the question is more likely to be correctly parsed.

(3) In terms of the intent detection problem, this paper mainly studies classification method based on CNN model and LSTM model. In this paper, traditional statistical learning method SVM and random forest are used as benchmark models to compare the effects of DAN model,CNN model and LSTM model. The experimental results show that the intent detection model based on CNN and LSTM model achieves better results than other algorithms. And the LSTM based model achieved the best results with Marco-F1 value of 92.57%.

(4) In terms of slot identification problem, this paper attempts the conditional random field model, BLSTM model, BLSTM-CRF model, BLSTM-CRF model with intent feature, and BLSTM-CRF model with intentional feature vocabulary. In addition, this paper attempts to identify the problem and the slot identification problem jointly, through the BLSTM-CRF model for joint learning. The experimental results show that the BLSTM-CRF model with the intent feature vocabulary has the

Abstract

best performance, with F1 value reaches 88.77.

(5)Finally, this paper completed the dialogue management module based on strategies, and completed the dialogue generation module in a template-based manner, and completed the construction of the task-oriented dialogue system for the travel domain.

Keywords: dialogue system,intent detection,slot filling,deep learning

摘要 .......................................................................................................................... I ABSTRACT................................................................................................................ II 第1章绪论 .. (1)

1.1课题背景及研究的目的和意义 (1)

1.2相关工作研究现状 (2)

1.2.1 基于模块化思想的对话系统研究 (2)

1.2.2 基于end-to-end的对话系统研究 (7)

1.3本文的主要研究内容 (8)

1.4论文主要组织架构 (9)

第2章面向出行领域的意图识别研究 (11)

2.1意图识别任务描述 (11)

2.2基于统计学习的意图识别模型 (11)

2.2.1 支持向量机介绍 (12)

2.2.2 随机森林介绍 (13)

2.2.3 特征选择 (13)

2.3基于深度学习的意图识别模型 (14)

2.3.1 DAN意图识别模型 (14)

2.3.2 CNN意图识别模型 (15)

2.3.3 LSTM意图识别模型 (16)

2.3.4 dropout正则化方法 (17)

2.3.5 批规范化方法 (18)

2.4意图识别实验 (18)

2.4.1 实验数据 (18)

2.4.2 评价指标 (19)

2.4.3 实验参数设置 (21)

2.4.4 实验结果对比分析 (22)

2.5本章小结 (24)

第3章面向出行领域的槽位识别问题研究 (25)

3.1槽位识别任务描述 (25)

3.2条件随机场介绍 (25)

3.2.1 条件随机场定义 (25)

3.2.2 条件随机场的参数化形式 (26)

3.2.4 条件随机场的相关问题 (26)

3.3基于BLSTM的槽位识别模型 (27)

3.3.1 基于BLSTM的槽位识别模型 (27)

3.3.2 基于BLSTM-CRF的槽位识别模型 (28)

3.3.3 动态RNN方法 (30)

3.4融合意图信息的槽位识别模型 (30)

3.4.1 BLSTM-CRF-INTENT槽位提取模型 (31)

3.4.2 BLSTM-CRF-INTENT_DICT槽位提取模型 (32)

3.4.3 意图识别和槽位识别联合学习模型 (34)

3.5槽位提取实验 (34)

3.5.1 实验数据 (34)

3.5.2 评价指标 (36)

3.5.3 实验参数设置 (37)

3.5.4 实验结果对比分析 (39)

3.6本章小结 (42)

第4章面向出行领域的多轮对话系统 (43)

4.1对话系统架构 (43)

4.2问句规范化 (44)

4.2.1 存在问题 (44)

4.2.2 问句规范化策略 (45)

4.3对话理解 (47)

4.4对话管理 (48)

4.4.1 对话状态跟踪 (48)

4.4.2 对话策略 (49)

4.5对话生成 (50)

4.6系统展示 (50)

4.7本章小结 (52)

结论 (53)

参考文献 (55)

攻读硕士学位期间发表的论文及其他成果 (60)

哈尔滨工业大学学位论文原创性声明和使用权限 (61)

致谢 (62)

附录 (63)