@inproceedings{vougiouklis-etal-2016-neural,
title = "A Neural Network Approach for Knowledge-Driven Response Generation",
author = "Vougiouklis, Pavlos and
Hare, Jonathon and
Simperl, Elena",
editor = "Matsumoto, Yuji and
Prasad, Rashmi",
booktitle = "Proceedings of {COLING} 2016, the 26th International Conference on Computational Linguistics: Technical Papers",
month = dec,
year = "2016",
address = "Osaka, Japan",
publisher = "The COLING 2016 Organizing Committee",
url = "https://aclanthology.org/C16-1318",
pages = "3370--3380",
abstract = "We present a novel response generation system. The system assumes the hypothesis that participants in a conversation base their response not only on previous dialog utterances but also on their background knowledge. Our model is based on a Recurrent Neural Network (RNN) that is trained over concatenated sequences of comments, a Convolution Neural Network that is trained over Wikipedia sentences and a formulation that couples the two trained embeddings in a multimodal space. We create a dataset of aligned Wikipedia sentences and sequences of Reddit utterances, which we we use to train our model. Given a sequence of past utterances and a set of sentences that represent the background knowledge, our end-to-end learnable model is able to generate context-sensitive and knowledge-driven responses by leveraging the alignment of two different data sources. Our approach achieves up to 55{\%} improvement in perplexity compared to purely sequential models based on RNNs that are trained only on sequences of utterances.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vougiouklis-etal-2016-neural">
<titleInfo>
<title>A Neural Network Approach for Knowledge-Driven Response Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pavlos</namePart>
<namePart type="family">Vougiouklis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathon</namePart>
<namePart type="family">Hare</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elena</namePart>
<namePart type="family">Simperl</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2016-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuji</namePart>
<namePart type="family">Matsumoto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rashmi</namePart>
<namePart type="family">Prasad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>The COLING 2016 Organizing Committee</publisher>
<place>
<placeTerm type="text">Osaka, Japan</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We present a novel response generation system. The system assumes the hypothesis that participants in a conversation base their response not only on previous dialog utterances but also on their background knowledge. Our model is based on a Recurrent Neural Network (RNN) that is trained over concatenated sequences of comments, a Convolution Neural Network that is trained over Wikipedia sentences and a formulation that couples the two trained embeddings in a multimodal space. We create a dataset of aligned Wikipedia sentences and sequences of Reddit utterances, which we we use to train our model. Given a sequence of past utterances and a set of sentences that represent the background knowledge, our end-to-end learnable model is able to generate context-sensitive and knowledge-driven responses by leveraging the alignment of two different data sources. Our approach achieves up to 55% improvement in perplexity compared to purely sequential models based on RNNs that are trained only on sequences of utterances.</abstract>
<identifier type="citekey">vougiouklis-etal-2016-neural</identifier>
<location>
<url>https://aclanthology.org/C16-1318</url>
</location>
<part>
<date>2016-12</date>
<extent unit="page">
<start>3370</start>
<end>3380</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Neural Network Approach for Knowledge-Driven Response Generation
%A Vougiouklis, Pavlos
%A Hare, Jonathon
%A Simperl, Elena
%Y Matsumoto, Yuji
%Y Prasad, Rashmi
%S Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers
%D 2016
%8 December
%I The COLING 2016 Organizing Committee
%C Osaka, Japan
%F vougiouklis-etal-2016-neural
%X We present a novel response generation system. The system assumes the hypothesis that participants in a conversation base their response not only on previous dialog utterances but also on their background knowledge. Our model is based on a Recurrent Neural Network (RNN) that is trained over concatenated sequences of comments, a Convolution Neural Network that is trained over Wikipedia sentences and a formulation that couples the two trained embeddings in a multimodal space. We create a dataset of aligned Wikipedia sentences and sequences of Reddit utterances, which we we use to train our model. Given a sequence of past utterances and a set of sentences that represent the background knowledge, our end-to-end learnable model is able to generate context-sensitive and knowledge-driven responses by leveraging the alignment of two different data sources. Our approach achieves up to 55% improvement in perplexity compared to purely sequential models based on RNNs that are trained only on sequences of utterances.
%U https://aclanthology.org/C16-1318
%P 3370-3380
Markdown (Informal)
[A Neural Network Approach for Knowledge-Driven Response Generation](https://aclanthology.org/C16-1318) (Vougiouklis et al., COLING 2016)
ACL