Deakin University
Browse
1/1
2 files

Neural-based natural language generation in dialogue using RNN encoder-decoder with semantic aggregation

conference contribution
posted on 2017-01-01, 00:00 authored by Van Khanh Tran, Le-Minh Nguyen, Satoshi Tojo
Natural language generation (NLG) is an
important component in spoken dialogue
systems. This paper presents a model
called Encoder-Aggregator-Decoder
which is an extension of an Recurrent
Neural Network based Encoder-Decoder
architecture. The proposed Semantic
Aggregator consists of two components:
an Aligner and a Refiner. The Aligner is
a conventional attention calculated over
the encoded input information, while
the Refiner is another attention or gating
mechanism stacked over the attentive
Aligner in order to further select and
aggregate the semantic elements. The
proposed model can be jointly trained
both text planning and text realization
to produce natural language utterances.
The model was extensively assessed on
four different NLG domains, in which the
results showed that the proposed generator consistently outperforms the previous
methods on all the NLG domains.

History

Event

Association for Computational Linguistics. Conference (18th : 2017 : Saarbrücken, Germany)

Series

Association for Computational Linguistics Conference

Pagination

231 - 240

Publisher

Association for Computational Linguistics

Location

Saarbrücken, Germany

Place of publication

Stroudsburg, Pa.

Start date

2017-08-15

End date

2017-08-17

ISBN-13

978-1-945626-82-1

Language

eng

Publication classification

E1.1 Full written paper - refereed

Copyright notice

2017, Association for Computational Linguistics

Editor/Contributor(s)

[Unknown]

Title of proceedings

SIGDIAL 2017 : Proceedings of the 18th Annual Meeting of the Special Interest Group on Discourse and Dialogue

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC