Unsupervised modeling of dialog acts in asynchronous conversations

Shafiq Rayhan Joty, Giuseppe Carenini, Chin Yew Lin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

26 Citations (Scopus)

Abstract

We present unsupervised approaches to the problem of modeling dialog acts in asynchronous conversations; i.e., conversations where participants collaborate with each other at different times. In particular, we investigate a graph-theoretic deterministic framework and two probabilistic conversation models (i.e., HMM and HMM+Mix) for modeling dialog acts in emails and forums. We train and test our conversation models on (a) temporal order and (b) graph-structural order of the datasets. Empirical evaluation suggests (i) the graph-theoretic framework that relies on lexical and structural similarity metrics is not the right model for this task, (ii) conversation models perform better on the graph-structural order than the temporal order of the datasets and (iii) HMM+Mix is a better conversation model than the simple HMM model.

Original languageEnglish
Title of host publicationIJCAI International Joint Conference on Artificial Intelligence
Pages1807-1813
Number of pages7
DOIs
Publication statusPublished - 1 Dec 2011
Externally publishedYes
Event22nd International Joint Conference on Artificial Intelligence, IJCAI 2011 - Barcelona, Catalonia, Spain
Duration: 16 Jul 201122 Jul 2011

Other

Other22nd International Joint Conference on Artificial Intelligence, IJCAI 2011
CountrySpain
CityBarcelona, Catalonia
Period16/7/1122/7/11

Fingerprint

Electronic mail
Statistical Models

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Rayhan Joty, S., Carenini, G., & Lin, C. Y. (2011). Unsupervised modeling of dialog acts in asynchronous conversations. In IJCAI International Joint Conference on Artificial Intelligence (pp. 1807-1813) https://doi.org/10.5591/978-1-57735-516-8/IJCAI11-303

Unsupervised modeling of dialog acts in asynchronous conversations. / Rayhan Joty, Shafiq; Carenini, Giuseppe; Lin, Chin Yew.

IJCAI International Joint Conference on Artificial Intelligence. 2011. p. 1807-1813.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Rayhan Joty, S, Carenini, G & Lin, CY 2011, Unsupervised modeling of dialog acts in asynchronous conversations. in IJCAI International Joint Conference on Artificial Intelligence. pp. 1807-1813, 22nd International Joint Conference on Artificial Intelligence, IJCAI 2011, Barcelona, Catalonia, Spain, 16/7/11. https://doi.org/10.5591/978-1-57735-516-8/IJCAI11-303
Rayhan Joty S, Carenini G, Lin CY. Unsupervised modeling of dialog acts in asynchronous conversations. In IJCAI International Joint Conference on Artificial Intelligence. 2011. p. 1807-1813 https://doi.org/10.5591/978-1-57735-516-8/IJCAI11-303
Rayhan Joty, Shafiq ; Carenini, Giuseppe ; Lin, Chin Yew. / Unsupervised modeling of dialog acts in asynchronous conversations. IJCAI International Joint Conference on Artificial Intelligence. 2011. pp. 1807-1813
@inproceedings{312c186444294ef1abb55865e7a7c3d2,
title = "Unsupervised modeling of dialog acts in asynchronous conversations",
abstract = "We present unsupervised approaches to the problem of modeling dialog acts in asynchronous conversations; i.e., conversations where participants collaborate with each other at different times. In particular, we investigate a graph-theoretic deterministic framework and two probabilistic conversation models (i.e., HMM and HMM+Mix) for modeling dialog acts in emails and forums. We train and test our conversation models on (a) temporal order and (b) graph-structural order of the datasets. Empirical evaluation suggests (i) the graph-theoretic framework that relies on lexical and structural similarity metrics is not the right model for this task, (ii) conversation models perform better on the graph-structural order than the temporal order of the datasets and (iii) HMM+Mix is a better conversation model than the simple HMM model.",
author = "{Rayhan Joty}, Shafiq and Giuseppe Carenini and Lin, {Chin Yew}",
year = "2011",
month = "12",
day = "1",
doi = "10.5591/978-1-57735-516-8/IJCAI11-303",
language = "English",
isbn = "9781577355120",
pages = "1807--1813",
booktitle = "IJCAI International Joint Conference on Artificial Intelligence",

}

TY - GEN

T1 - Unsupervised modeling of dialog acts in asynchronous conversations

AU - Rayhan Joty, Shafiq

AU - Carenini, Giuseppe

AU - Lin, Chin Yew

PY - 2011/12/1

Y1 - 2011/12/1

N2 - We present unsupervised approaches to the problem of modeling dialog acts in asynchronous conversations; i.e., conversations where participants collaborate with each other at different times. In particular, we investigate a graph-theoretic deterministic framework and two probabilistic conversation models (i.e., HMM and HMM+Mix) for modeling dialog acts in emails and forums. We train and test our conversation models on (a) temporal order and (b) graph-structural order of the datasets. Empirical evaluation suggests (i) the graph-theoretic framework that relies on lexical and structural similarity metrics is not the right model for this task, (ii) conversation models perform better on the graph-structural order than the temporal order of the datasets and (iii) HMM+Mix is a better conversation model than the simple HMM model.

AB - We present unsupervised approaches to the problem of modeling dialog acts in asynchronous conversations; i.e., conversations where participants collaborate with each other at different times. In particular, we investigate a graph-theoretic deterministic framework and two probabilistic conversation models (i.e., HMM and HMM+Mix) for modeling dialog acts in emails and forums. We train and test our conversation models on (a) temporal order and (b) graph-structural order of the datasets. Empirical evaluation suggests (i) the graph-theoretic framework that relies on lexical and structural similarity metrics is not the right model for this task, (ii) conversation models perform better on the graph-structural order than the temporal order of the datasets and (iii) HMM+Mix is a better conversation model than the simple HMM model.

UR - http://www.scopus.com/inward/record.url?scp=84876815928&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876815928&partnerID=8YFLogxK

U2 - 10.5591/978-1-57735-516-8/IJCAI11-303

DO - 10.5591/978-1-57735-516-8/IJCAI11-303

M3 - Conference contribution

AN - SCOPUS:84876815928

SN - 9781577355120

SP - 1807

EP - 1813

BT - IJCAI International Joint Conference on Artificial Intelligence

ER -