Five-way Smoking Status Classification Using Text Hot-Spot Identification and Error-correcting Output Codes

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

We participated in the i2b2 smoking status classification challenge task. The purpose of this task was to evaluate the ability of systems to automatically identify patient smoking status from discharge summaries. Our submission included several techniques that we compared and studied, including hot-spot identification, zero-vector filtering, inverse class frequency weighting, error-correcting output codes, and post-processing rules. We evaluated our approaches using the same methods as the i2b2 task organizers, using micro- and macro-averaged F1 as the primary performance metric. Our best performing system achieved a micro-F1 of 0.9000 on the test collection, equivalent to the best performing system submitted to the i2b2 challenge. Hot-spot identification, zero-vector filtering, classifier weighting, and error correcting output coding contributed additively to increased performance, with hot-spot identification having by far the largest positive effect. High performance on automatic identification of patient smoking status from discharge summaries is achievable with the efficient and straightforward machine learning techniques studied here.

Original languageEnglish (US)
Pages (from-to)32-35
Number of pages4
JournalJournal of the American Medical Informatics Association
Volume15
Issue number1
DOIs
StatePublished - Jan 2008

Fingerprint

Smoking
Aptitude
Machine Learning

ASJC Scopus subject areas

  • Medicine(all)

Cite this

@article{e5c646d6b0214103a1aa42f404066ef4,
title = "Five-way Smoking Status Classification Using Text Hot-Spot Identification and Error-correcting Output Codes",
abstract = "We participated in the i2b2 smoking status classification challenge task. The purpose of this task was to evaluate the ability of systems to automatically identify patient smoking status from discharge summaries. Our submission included several techniques that we compared and studied, including hot-spot identification, zero-vector filtering, inverse class frequency weighting, error-correcting output codes, and post-processing rules. We evaluated our approaches using the same methods as the i2b2 task organizers, using micro- and macro-averaged F1 as the primary performance metric. Our best performing system achieved a micro-F1 of 0.9000 on the test collection, equivalent to the best performing system submitted to the i2b2 challenge. Hot-spot identification, zero-vector filtering, classifier weighting, and error correcting output coding contributed additively to increased performance, with hot-spot identification having by far the largest positive effect. High performance on automatic identification of patient smoking status from discharge summaries is achievable with the efficient and straightforward machine learning techniques studied here.",
author = "Aaron Cohen",
year = "2008",
month = "1",
doi = "10.1197/jamia.M2434",
language = "English (US)",
volume = "15",
pages = "32--35",
journal = "Journal of the American Medical Informatics Association",
issn = "1067-5027",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - Five-way Smoking Status Classification Using Text Hot-Spot Identification and Error-correcting Output Codes

AU - Cohen, Aaron

PY - 2008/1

Y1 - 2008/1

N2 - We participated in the i2b2 smoking status classification challenge task. The purpose of this task was to evaluate the ability of systems to automatically identify patient smoking status from discharge summaries. Our submission included several techniques that we compared and studied, including hot-spot identification, zero-vector filtering, inverse class frequency weighting, error-correcting output codes, and post-processing rules. We evaluated our approaches using the same methods as the i2b2 task organizers, using micro- and macro-averaged F1 as the primary performance metric. Our best performing system achieved a micro-F1 of 0.9000 on the test collection, equivalent to the best performing system submitted to the i2b2 challenge. Hot-spot identification, zero-vector filtering, classifier weighting, and error correcting output coding contributed additively to increased performance, with hot-spot identification having by far the largest positive effect. High performance on automatic identification of patient smoking status from discharge summaries is achievable with the efficient and straightforward machine learning techniques studied here.

AB - We participated in the i2b2 smoking status classification challenge task. The purpose of this task was to evaluate the ability of systems to automatically identify patient smoking status from discharge summaries. Our submission included several techniques that we compared and studied, including hot-spot identification, zero-vector filtering, inverse class frequency weighting, error-correcting output codes, and post-processing rules. We evaluated our approaches using the same methods as the i2b2 task organizers, using micro- and macro-averaged F1 as the primary performance metric. Our best performing system achieved a micro-F1 of 0.9000 on the test collection, equivalent to the best performing system submitted to the i2b2 challenge. Hot-spot identification, zero-vector filtering, classifier weighting, and error correcting output coding contributed additively to increased performance, with hot-spot identification having by far the largest positive effect. High performance on automatic identification of patient smoking status from discharge summaries is achievable with the efficient and straightforward machine learning techniques studied here.

UR - http://www.scopus.com/inward/record.url?scp=36749057750&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36749057750&partnerID=8YFLogxK

U2 - 10.1197/jamia.M2434

DO - 10.1197/jamia.M2434

M3 - Article

C2 - 17947623

AN - SCOPUS:36749057750

VL - 15

SP - 32

EP - 35

JO - Journal of the American Medical Informatics Association

JF - Journal of the American Medical Informatics Association

SN - 1067-5027

IS - 1

ER -