Cargando…

Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes

In 2007, suicide was the tenth leading cause of death in the U.S. Given the significance of this problem, suicide was the focus of the 2011 Informatics for Integrating Biology and the Bedside (i2b2) Natural Language Processing (NLP) shared task competition (track two). Specifically, the challenge co...

Descripción completa

Detalles Bibliográficos
Autores principales: McCart, James A., Finch, Dezon K., Jarman, Jay, Hickling, Edward, Lind, Jason D., Richardson, Matthew R., Berndt, Donald J., Luther, Stephen L.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Libertas Academica 2012
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3409473/
https://www.ncbi.nlm.nih.gov/pubmed/22879763
http://dx.doi.org/10.4137/BII.S8931
_version_ 1782239590572097536
author McCart, James A.
Finch, Dezon K.
Jarman, Jay
Hickling, Edward
Lind, Jason D.
Richardson, Matthew R.
Berndt, Donald J.
Luther, Stephen L.
author_facet McCart, James A.
Finch, Dezon K.
Jarman, Jay
Hickling, Edward
Lind, Jason D.
Richardson, Matthew R.
Berndt, Donald J.
Luther, Stephen L.
author_sort McCart, James A.
collection PubMed
description In 2007, suicide was the tenth leading cause of death in the U.S. Given the significance of this problem, suicide was the focus of the 2011 Informatics for Integrating Biology and the Bedside (i2b2) Natural Language Processing (NLP) shared task competition (track two). Specifically, the challenge concentrated on sentiment analysis, predicting the presence or absence of 15 emotions (labels) simultaneously in a collection of suicide notes spanning over 70 years. Our team explored multiple approaches combining regular expression-based rules, statistical text mining (STM), and an approach that applies weights to text while accounting for multiple labels. Our best submission used an ensemble of both rules and STM models to achieve a micro-averaged F(1) score of 0.5023, slightly above the mean from the 26 teams that competed (0.4875).
format Online
Article
Text
id pubmed-3409473
institution National Center for Biotechnology Information
language English
publishDate 2012
publisher Libertas Academica
record_format MEDLINE/PubMed
spelling pubmed-34094732012-08-09 Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes McCart, James A. Finch, Dezon K. Jarman, Jay Hickling, Edward Lind, Jason D. Richardson, Matthew R. Berndt, Donald J. Luther, Stephen L. Biomed Inform Insights Original Research In 2007, suicide was the tenth leading cause of death in the U.S. Given the significance of this problem, suicide was the focus of the 2011 Informatics for Integrating Biology and the Bedside (i2b2) Natural Language Processing (NLP) shared task competition (track two). Specifically, the challenge concentrated on sentiment analysis, predicting the presence or absence of 15 emotions (labels) simultaneously in a collection of suicide notes spanning over 70 years. Our team explored multiple approaches combining regular expression-based rules, statistical text mining (STM), and an approach that applies weights to text while accounting for multiple labels. Our best submission used an ensemble of both rules and STM models to achieve a micro-averaged F(1) score of 0.5023, slightly above the mean from the 26 teams that competed (0.4875). Libertas Academica 2012-01-30 /pmc/articles/PMC3409473/ /pubmed/22879763 http://dx.doi.org/10.4137/BII.S8931 Text en © the author(s), publisher and licensee Libertas Academica Ltd. This is an open access article. Unrestricted non-commercial use is permitted provided the original work is properly cited.
spellingShingle Original Research
McCart, James A.
Finch, Dezon K.
Jarman, Jay
Hickling, Edward
Lind, Jason D.
Richardson, Matthew R.
Berndt, Donald J.
Luther, Stephen L.
Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
title Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
title_full Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
title_fullStr Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
title_full_unstemmed Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
title_short Using Ensemble Models to Classify the Sentiment Expressed in Suicide Notes
title_sort using ensemble models to classify the sentiment expressed in suicide notes
topic Original Research
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3409473/
https://www.ncbi.nlm.nih.gov/pubmed/22879763
http://dx.doi.org/10.4137/BII.S8931
work_keys_str_mv AT mccartjamesa usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT finchdezonk usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT jarmanjay usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT hicklingedward usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT lindjasond usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT richardsonmatthewr usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT berndtdonaldj usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes
AT lutherstephenl usingensemblemodelstoclassifythesentimentexpressedinsuicidenotes