Pattern-based system to extract and distinguish drugadverse effect relation from other drug-medical condition relations in the same sentence
Extraction of drug-adverse effect causal relationship supports pharmacovigilance research and reduces the manual efforts for some tasks such as drug safety monitoring and building databases for adverse drugs effects from free text. In this study, we proposed a pattern-based method to extract drug-ad...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Asian Research Publishing Network
2015
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/58765/ http://www.arpnjournals.com/jeas/volume_03_2015.htm |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Teknologi Malaysia |
| Summary: | Extraction of drug-adverse effect causal relationship supports pharmacovigilance research and reduces the manual efforts for some tasks such as drug safety monitoring and building databases for adverse drugs effects from free text. In this study, we proposed a pattern-based method to extract drug-adverse effects causal relation from medical case reports and to distinguish this relation from other drug-medical condition relations exist in the same sentences. For training and evaluation purposes; we used 481 sentences from ADE corpus. Our method combined a concept recognition system with a module for drug-adverse effect relation extraction and discrimination task based on automatic generated numerous patterns and the position of matched pattern in a sentence. Our method achieved recall of 36.1, precision of 30.6 and an F-Score of 33.1 .The result of this study provides rapid extraction of machine-understandable drug-adverse effects pairs which can help in many computational drug researches. |
|---|