@inproceedings{du-etal-2023-two,
title = "A Two-Stage Progressive Intent Clustering for Task-Oriented Dialogue",
author = "Du, Bingzhu and
Su, Nan and
Zhang, Yuchi and
Wang, Yongliang",
editor = "Chen, Yun-Nung and
Crook, Paul and
Galley, Michel and
Ghazarian, Sarik and
Gunasekara, Chulaka and
Gupta, Raghav and
Hedayatnia, Behnam and
Kottur, Satwik and
Moon, Seungwhan and
Zhang, Chen",
booktitle = "Proceedings of The Eleventh Dialog System Technology Challenge",
month = sep,
year = "2023",
address = "Prague, Czech Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.dstc-1.7",
pages = "48--56",
abstract = "Natural Language Understanding (NLU) is one of the most critical components of task-oriented dialogue, and it is often considered as an intent classification task. To achieve outstanding intent identification performance, system designers often need to hire a large number of domain experts to label the data, which is inefficient and costly. To address this problem, researchers{'} attention has gradually shifted to automatic intent clustering methods, which employ low-resource unsupervised approaches to solve classification problems. The classical framework for clustering is deep clustering, which uses deep neural networks (DNNs) to jointly optimize non-clustering loss and clustering loss. However, for new conversational domains or services, utterances required to assign intents are scarce and the performance of DNNs is often dependent on large amounts of data. In addition, although re-clustering with k-means algorithm after training the network usually leads to better results, k-means methods often suffer from poor stability. To address these problems, we propose an effective two-stage progressive approach to refine the clustering. Firstly, we pre-train the network with contrastive loss using all conversations data and then optimize the clustering loss and contrastive loss simultaneously. Secondly, we propose adaptive progressive k-means to alleviate the randomness of vanilla k-means, achieving better performance and smaller deviation. Our method ranks second in DSTC11 Track2 Task 1, a benchmark for intent clustering of task-oriented dialogue, demonstrating the superiority and effectiveness of our method.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="du-etal-2023-two">
<titleInfo>
<title>A Two-Stage Progressive Intent Clustering for Task-Oriented Dialogue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bingzhu</namePart>
<namePart type="family">Du</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nan</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuchi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yongliang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of The Eleventh Dialog System Technology Challenge</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">Crook</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michel</namePart>
<namePart type="family">Galley</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sarik</namePart>
<namePart type="family">Ghazarian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chulaka</namePart>
<namePart type="family">Gunasekara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raghav</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Behnam</namePart>
<namePart type="family">Hedayatnia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Satwik</namePart>
<namePart type="family">Kottur</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seungwhan</namePart>
<namePart type="family">Moon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Prague, Czech Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Natural Language Understanding (NLU) is one of the most critical components of task-oriented dialogue, and it is often considered as an intent classification task. To achieve outstanding intent identification performance, system designers often need to hire a large number of domain experts to label the data, which is inefficient and costly. To address this problem, researchers’ attention has gradually shifted to automatic intent clustering methods, which employ low-resource unsupervised approaches to solve classification problems. The classical framework for clustering is deep clustering, which uses deep neural networks (DNNs) to jointly optimize non-clustering loss and clustering loss. However, for new conversational domains or services, utterances required to assign intents are scarce and the performance of DNNs is often dependent on large amounts of data. In addition, although re-clustering with k-means algorithm after training the network usually leads to better results, k-means methods often suffer from poor stability. To address these problems, we propose an effective two-stage progressive approach to refine the clustering. Firstly, we pre-train the network with contrastive loss using all conversations data and then optimize the clustering loss and contrastive loss simultaneously. Secondly, we propose adaptive progressive k-means to alleviate the randomness of vanilla k-means, achieving better performance and smaller deviation. Our method ranks second in DSTC11 Track2 Task 1, a benchmark for intent clustering of task-oriented dialogue, demonstrating the superiority and effectiveness of our method.</abstract>
<identifier type="citekey">du-etal-2023-two</identifier>
<location>
<url>https://aclanthology.org/2023.dstc-1.7</url>
</location>
<part>
<date>2023-09</date>
<extent unit="page">
<start>48</start>
<end>56</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Two-Stage Progressive Intent Clustering for Task-Oriented Dialogue
%A Du, Bingzhu
%A Su, Nan
%A Zhang, Yuchi
%A Wang, Yongliang
%Y Chen, Yun-Nung
%Y Crook, Paul
%Y Galley, Michel
%Y Ghazarian, Sarik
%Y Gunasekara, Chulaka
%Y Gupta, Raghav
%Y Hedayatnia, Behnam
%Y Kottur, Satwik
%Y Moon, Seungwhan
%Y Zhang, Chen
%S Proceedings of The Eleventh Dialog System Technology Challenge
%D 2023
%8 September
%I Association for Computational Linguistics
%C Prague, Czech Republic
%F du-etal-2023-two
%X Natural Language Understanding (NLU) is one of the most critical components of task-oriented dialogue, and it is often considered as an intent classification task. To achieve outstanding intent identification performance, system designers often need to hire a large number of domain experts to label the data, which is inefficient and costly. To address this problem, researchers’ attention has gradually shifted to automatic intent clustering methods, which employ low-resource unsupervised approaches to solve classification problems. The classical framework for clustering is deep clustering, which uses deep neural networks (DNNs) to jointly optimize non-clustering loss and clustering loss. However, for new conversational domains or services, utterances required to assign intents are scarce and the performance of DNNs is often dependent on large amounts of data. In addition, although re-clustering with k-means algorithm after training the network usually leads to better results, k-means methods often suffer from poor stability. To address these problems, we propose an effective two-stage progressive approach to refine the clustering. Firstly, we pre-train the network with contrastive loss using all conversations data and then optimize the clustering loss and contrastive loss simultaneously. Secondly, we propose adaptive progressive k-means to alleviate the randomness of vanilla k-means, achieving better performance and smaller deviation. Our method ranks second in DSTC11 Track2 Task 1, a benchmark for intent clustering of task-oriented dialogue, demonstrating the superiority and effectiveness of our method.
%U https://aclanthology.org/2023.dstc-1.7
%P 48-56
Markdown (Informal)
[A Two-Stage Progressive Intent Clustering for Task-Oriented Dialogue](https://aclanthology.org/2023.dstc-1.7) (Du et al., DSTC-WS 2023)
ACL