Hybrid system for adverse drug event detection

Alec B. Chapman, Kelly S. Peterson, Patrick R. Alba, Scott L. DuVall, Olga V. Patterson
Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection, PMLR 90:16-24, 2018.

Abstract

In context of the NLP Challenges for Detecting Medication and Adverse Drug Events from Electronic Health Records (MADE 1.0) \citep{Yu}, we built a hybrid natural language processing system that combined multiple algorithms and resources to identify the relationship between mentions of symptoms and drugs. Our system employed a conditional random field (CRF) model for named entity recognition (NER) and a random forest model for relation extraction (RE). Final performance of each model was evaluated separately and then combined on the challenge’s hold-out evaluation set. The micro-averaged F1 score was 80.9% for NER, 86.8% for RE, and 59.2% for the final system.

Cite this Paper

BibTeX
@InProceedings{pmlr-v90-chapman18a, title = {Hybrid system for adverse drug event detection}, author = {Chapman, Alec B. and Peterson, Kelly S. and Alba, Patrick R. and DuVall, Scott L. and Patterson, Olga V.}, booktitle = {Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection}, pages = {16--24}, year = {2018}, editor = {Liu, Feifan and Jagannatha, Abhyuday and Yu, Hong}, volume = {90}, series = {Proceedings of Machine Learning Research}, month = {04 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v90/chapman18a/chapman18a.pdf}, url = {https://proceedings.mlr.press/v90/chapman18a.html}, abstract = {In context of the NLP Challenges for Detecting Medication and Adverse Drug Events from Electronic Health Records (MADE 1.0) \citep{Yu}, we built a hybrid natural language processing system that combined multiple algorithms and resources to identify the relationship between mentions of symptoms and drugs. Our system employed a conditional random field (CRF) model for named entity recognition (NER) and a random forest model for relation extraction (RE). Final performance of each model was evaluated separately and then combined on the challenge’s hold-out evaluation set. The micro-averaged F1 score was 80.9% for NER, 86.8% for RE, and 59.2% for the final system.} }
Endnote
%0 Conference Paper %T Hybrid system for adverse drug event detection %A Alec B. Chapman %A Kelly S. Peterson %A Patrick R. Alba %A Scott L. DuVall %A Olga V. Patterson %B Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection %C Proceedings of Machine Learning Research %D 2018 %E Feifan Liu %E Abhyuday Jagannatha %E Hong Yu %F pmlr-v90-chapman18a %I PMLR %P 16--24 %U https://proceedings.mlr.press/v90/chapman18a.html %V 90 %X In context of the NLP Challenges for Detecting Medication and Adverse Drug Events from Electronic Health Records (MADE 1.0) \citep{Yu}, we built a hybrid natural language processing system that combined multiple algorithms and resources to identify the relationship between mentions of symptoms and drugs. Our system employed a conditional random field (CRF) model for named entity recognition (NER) and a random forest model for relation extraction (RE). Final performance of each model was evaluated separately and then combined on the challenge’s hold-out evaluation set. The micro-averaged F1 score was 80.9% for NER, 86.8% for RE, and 59.2% for the final system.
APA
Chapman, A.B., Peterson, K.S., Alba, P.R., DuVall, S.L. & Patterson, O.V.. (2018). Hybrid system for adverse drug event detection. Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection, in Proceedings of Machine Learning Research 90:16-24 Available from https://proceedings.mlr.press/v90/chapman18a.html.

Related Material