Skip to main navigation Skip to search Skip to main content

Learning Personalized End-to-End Task-Oriented Dialogue Generation

  • School of Computer Science and Technology, Harbin Institute of Technology
  • Shenzhen University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Building personalized task-oriented dialogue system is an important but challenging task. Significant success has been achieved by selecting the responses from the pre-defined template. However, preparing massive response template is time-consuming and human-labor intensive. In this paper, we propose an end-to-end framework based on the memory networks for responses generation in the personalized task-oriented dialog system. The static attention mechanism is used to encode the user-conversation relationship to form a global vector representation, and the dynamic attention mechanism is used to obtain import local information during the decoding phase. In addition, we propose a gating mechanism to incorporate user information into the network to enhance the personalized ability of the response. Experiments on the benchmark dataset show that our model achieves better performance than the strong baseline methods in personalized task-oriented dialogue generation.

Original languageEnglish
Title of host publicationNatural Language Processing and Chinese Computing - 8th CCF International Conference, NLPCC 2019, Proceedings
EditorsJie Tang, Min-Yen Kan, Dongyan Zhao, Sujian Li, Hongying Zan
PublisherSpringer
Pages55-66
Number of pages12
ISBN (Print)9783030322328
DOIs
StatePublished - 2019
Externally publishedYes
Event8th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2019 - Dunhuang, China
Duration: 9 Oct 201914 Oct 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11838 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference8th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2019
Country/TerritoryChina
CityDunhuang
Period9/10/1914/10/19

Keywords

  • Dialogue generation
  • Personalized response
  • Task-oriented dialogue system

Fingerprint

Dive into the research topics of 'Learning Personalized End-to-End Task-Oriented Dialogue Generation'. Together they form a unique fingerprint.

Cite this