On the collective classification of email "speech acts"
V. Carvalho, and W. Cohen. SIGIR '05: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval, page 345--352. New York, NY, USA, ACM, (2005)
DOI: 10.1145/1076034.1076094
Abstract
We consider classification of email messages as to whether or not they contain certain "email acts", such as a request or a commitment. We show that exploiting the sequential correlation among email messages in the same thread can improve email-act classification. More specifically, we describe a new text-classification algorithm based on a dependency-network based collective classification method, in which the local classifiers are maximum entropy models based on words and certain relational features. We show that statistically significant improvements over a bag-of-words baseline classifier can be obtained for some, but not all, email-act classes. Performance improvements obtained by collective classification appears to be consistent across many email acts suggested by prior speech-act theory.
Description
On the collective classification of email "speech acts"
%0 Conference Paper
%1 Carvalho05
%A Carvalho, Vitor R.
%A Cohen, William W.
%B SIGIR '05: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval
%C New York, NY, USA
%D 2005
%I ACM
%K intent
%P 345--352
%R 10.1145/1076034.1076094
%T On the collective classification of email "speech acts"
%U http://portal.acm.org/citation.cfm?doid=1076034.1076094
%X We consider classification of email messages as to whether or not they contain certain "email acts", such as a request or a commitment. We show that exploiting the sequential correlation among email messages in the same thread can improve email-act classification. More specifically, we describe a new text-classification algorithm based on a dependency-network based collective classification method, in which the local classifiers are maximum entropy models based on words and certain relational features. We show that statistically significant improvements over a bag-of-words baseline classifier can be obtained for some, but not all, email-act classes. Performance improvements obtained by collective classification appears to be consistent across many email acts suggested by prior speech-act theory.
%@ 1-59593-034-5
@inproceedings{Carvalho05,
abstract = {We consider classification of email messages as to whether or not they contain certain "email acts", such as a request or a commitment. We show that exploiting the sequential correlation among email messages in the same thread can improve email-act classification. More specifically, we describe a new text-classification algorithm based on a dependency-network based collective classification method, in which the local classifiers are maximum entropy models based on words and certain relational features. We show that statistically significant improvements over a bag-of-words baseline classifier can be obtained for some, but not all, email-act classes. Performance improvements obtained by collective classification appears to be consistent across many email acts suggested by prior speech-act theory.},
added-at = {2010-05-07T11:13:51.000+0200},
address = {New York, NY, USA},
author = {Carvalho, Vitor R. and Cohen, William W.},
biburl = {https://www.bibsonomy.org/bibtex/22d920c5d8fe0458791a051646a93f83b/mkroell},
booktitle = {SIGIR '05: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval},
description = {On the collective classification of email "speech acts"},
doi = {10.1145/1076034.1076094},
interhash = {60b41305d48f7b756219d56d0396d200},
intrahash = {2d920c5d8fe0458791a051646a93f83b},
isbn = {1-59593-034-5},
keywords = {intent},
location = {Salvador, Brazil},
pages = {345--352},
publisher = {ACM},
timestamp = {2010-05-07T11:13:51.000+0200},
title = {On the collective classification of email "speech acts"},
url = {http://portal.acm.org/citation.cfm?doid=1076034.1076094},
year = 2005
}