In this work, we propose a new statistical model for building robust dialog systems using neural networks to either retrieve or generate dialog response based on an existing data sources. In the retrieval task, we propose an approach that uses paraphrase identification during the retrieval process. This is done by employing recursive autoencoders and dynamic pooling to determine whether two sentences with arbitrary length have the same meaning. For both the generation and retrieval tasks, we propose a model using long short term memory (LSTM) neural networks that works by first using an LSTM encoder to read in the user's utterance into a continuous vector-space representation, then using an LSTM decoder to generate the most probable word sequence. An evaluation based on objective and subjective metrics shows that the new proposed approaches have the ability to deal with user inputs that are not well covered in the database compared to standard example-based dialog baselines.
Lasguido NIO
Nara Institute of Science and Technology
Sakriani SAKTI
Nara Institute of Science and Technology
Graham NEUBIG
Nara Institute of Science and Technology
Koichiro YOSHINO
Nara Institute of Science and Technology
Satoshi NAKAMURA
Nara Institute of Science and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Lasguido NIO, Sakriani SAKTI, Graham NEUBIG, Koichiro YOSHINO, Satoshi NAKAMURA, "Neural Network Approaches to Dialog Response Retrieval and Generation" in IEICE TRANSACTIONS on Information,
vol. E99-D, no. 10, pp. 2508-2517, October 2016, doi: 10.1587/transinf.2016SLP0018.
Abstract: In this work, we propose a new statistical model for building robust dialog systems using neural networks to either retrieve or generate dialog response based on an existing data sources. In the retrieval task, we propose an approach that uses paraphrase identification during the retrieval process. This is done by employing recursive autoencoders and dynamic pooling to determine whether two sentences with arbitrary length have the same meaning. For both the generation and retrieval tasks, we propose a model using long short term memory (LSTM) neural networks that works by first using an LSTM encoder to read in the user's utterance into a continuous vector-space representation, then using an LSTM decoder to generate the most probable word sequence. An evaluation based on objective and subjective metrics shows that the new proposed approaches have the ability to deal with user inputs that are not well covered in the database compared to standard example-based dialog baselines.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2016SLP0018/_p
Copy
@ARTICLE{e99-d_10_2508,
author={Lasguido NIO, Sakriani SAKTI, Graham NEUBIG, Koichiro YOSHINO, Satoshi NAKAMURA, },
journal={IEICE TRANSACTIONS on Information},
title={Neural Network Approaches to Dialog Response Retrieval and Generation},
year={2016},
volume={E99-D},
number={10},
pages={2508-2517},
abstract={In this work, we propose a new statistical model for building robust dialog systems using neural networks to either retrieve or generate dialog response based on an existing data sources. In the retrieval task, we propose an approach that uses paraphrase identification during the retrieval process. This is done by employing recursive autoencoders and dynamic pooling to determine whether two sentences with arbitrary length have the same meaning. For both the generation and retrieval tasks, we propose a model using long short term memory (LSTM) neural networks that works by first using an LSTM encoder to read in the user's utterance into a continuous vector-space representation, then using an LSTM decoder to generate the most probable word sequence. An evaluation based on objective and subjective metrics shows that the new proposed approaches have the ability to deal with user inputs that are not well covered in the database compared to standard example-based dialog baselines.},
keywords={},
doi={10.1587/transinf.2016SLP0018},
ISSN={1745-1361},
month={October},}
Copy
TY - JOUR
TI - Neural Network Approaches to Dialog Response Retrieval and Generation
T2 - IEICE TRANSACTIONS on Information
SP - 2508
EP - 2517
AU - Lasguido NIO
AU - Sakriani SAKTI
AU - Graham NEUBIG
AU - Koichiro YOSHINO
AU - Satoshi NAKAMURA
PY - 2016
DO - 10.1587/transinf.2016SLP0018
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E99-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2016
AB - In this work, we propose a new statistical model for building robust dialog systems using neural networks to either retrieve or generate dialog response based on an existing data sources. In the retrieval task, we propose an approach that uses paraphrase identification during the retrieval process. This is done by employing recursive autoencoders and dynamic pooling to determine whether two sentences with arbitrary length have the same meaning. For both the generation and retrieval tasks, we propose a model using long short term memory (LSTM) neural networks that works by first using an LSTM encoder to read in the user's utterance into a continuous vector-space representation, then using an LSTM decoder to generate the most probable word sequence. An evaluation based on objective and subjective metrics shows that the new proposed approaches have the ability to deal with user inputs that are not well covered in the database compared to standard example-based dialog baselines.
ER -