UArizona at the MADE1.0 NLP Challenge

Dongfang Xu; Vikas Yadav; Steven Bethard

UArizona at the MADE1.0 NLP Challenge

Dongfang Xu, Vikas Yadav, Steven Bethard

Information, School of

Research output: Contribution to journal › Conference article › peer-review

15 Scopus citations

Abstract

MADE1.0 is a public natural language processing challenge aiming to extract medication and adverse drug events from Electronic Health Records. This work presents NER and RI systems developed by UArizona team for the MADE1.0 competition. We propose a neural NER system for medical named entity recognition using both local and context features for each individual word and a simple but effective SVM-based pairwise relation classification system for identifying relations between medical entities and attributes. Our system achieves 81.56%, 83.18%, and 59.85% F1 score in the three tasks of MADE1.0 challenge, respectively, ranked amongst the top three teams for Task 2 and 3.

Original language	English (US)
Pages (from-to)	57-65
Number of pages	9
Journal	Proceedings of Machine Learning Research
Volume	90
State	Published - 2018
Event	1st International Workshop on Medication and Adverse Drug Event Detection, MADE 2018 - Duration: May 4 2018 → …

Keywords

Adverse Drug Event
Information Extraction
Neural Network

ASJC Scopus subject areas

Artificial Intelligence
Software
Control and Systems Engineering
Statistics and Probability

Cite this

@article{e3bea6cfcdeb40a48e198878b8412f77,

title = "UArizona at the MADE1.0 NLP Challenge",

abstract = "MADE1.0 is a public natural language processing challenge aiming to extract medication and adverse drug events from Electronic Health Records. This work presents NER and RI systems developed by UArizona team for the MADE1.0 competition. We propose a neural NER system for medical named entity recognition using both local and context features for each individual word and a simple but effective SVM-based pairwise relation classification system for identifying relations between medical entities and attributes. Our system achieves 81.56%, 83.18%, and 59.85% F1 score in the three tasks of MADE1.0 challenge, respectively, ranked amongst the top three teams for Task 2 and 3.",

keywords = "Adverse Drug Event, Information Extraction, Neural Network",

author = "Dongfang Xu and Vikas Yadav and Steven Bethard",

note = "Publisher Copyright: {\textcopyright} 2018 D. Xu, V. Yadav∗ & S. Bethard.; 1st International Workshop on Medication and Adverse Drug Event Detection, MADE 2018 ; Conference date: 04-05-2018",

year = "2018",

language = "English (US)",

volume = "90",

pages = "57--65",

journal = "Proceedings of Machine Learning Research",

issn = "2640-3498",

publisher = "ML Research Press",

}

TY - JOUR

T1 - UArizona at the MADE1.0 NLP Challenge

AU - Xu, Dongfang

AU - Yadav, Vikas

AU - Bethard, Steven

PY - 2018

Y1 - 2018

N2 - MADE1.0 is a public natural language processing challenge aiming to extract medication and adverse drug events from Electronic Health Records. This work presents NER and RI systems developed by UArizona team for the MADE1.0 competition. We propose a neural NER system for medical named entity recognition using both local and context features for each individual word and a simple but effective SVM-based pairwise relation classification system for identifying relations between medical entities and attributes. Our system achieves 81.56%, 83.18%, and 59.85% F1 score in the three tasks of MADE1.0 challenge, respectively, ranked amongst the top three teams for Task 2 and 3.

AB - MADE1.0 is a public natural language processing challenge aiming to extract medication and adverse drug events from Electronic Health Records. This work presents NER and RI systems developed by UArizona team for the MADE1.0 competition. We propose a neural NER system for medical named entity recognition using both local and context features for each individual word and a simple but effective SVM-based pairwise relation classification system for identifying relations between medical entities and attributes. Our system achieves 81.56%, 83.18%, and 59.85% F1 score in the three tasks of MADE1.0 challenge, respectively, ranked amongst the top three teams for Task 2 and 3.

KW - Adverse Drug Event

KW - Information Extraction

KW - Neural Network

UR - http://www.scopus.com/inward/record.url?scp=85100495643&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85100495643&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85100495643

SN - 2640-3498

VL - 90

SP - 57

EP - 65

JO - Proceedings of Machine Learning Research

JF - Proceedings of Machine Learning Research

T2 - 1st International Workshop on Medication and Adverse Drug Event Detection, MADE 2018

Y2 - 4 May 2018

ER -

UArizona at the MADE1.0 NLP Challenge

Abstract

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this