A Novel Feature Selection Strategy for Enhanced Biomedical Event Extraction Using the Turku System
Joint Authors
Fang, Alex Chengyu
Xia, Jingbo
Zhang, Xing
Source
Issue
Vol. 2014, Issue 2014 (31 Dec. 2014), pp.1-12, 12 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2014-04-06
Country of Publication
Egypt
No. of Pages
12
Main Subjects
Abstract EN
Feature selection is of paramount importance for text-mining classifiers with high-dimensional features.
The Turku Event Extraction System (TEES) is the best performing tool in the GENIA BioNLP 2009/2011 shared tasks, which relies heavily on high-dimensional features.
This paper describes research which, based on an implementation of an accumulated effect evaluation (AEE) algorithm applying the greedy search strategy, analyses the contribution of every single feature class in TEES with a view to identify important features and modify the feature set accordingly.
With an updated feature set, a new system is acquired with enhanced performance which achieves an increased F-score of 53.27% up from 51.21% for Task 1 under strict evaluation criteria and 57.24% according to the approximate span and recursive criterion.
American Psychological Association (APA)
Xia, Jingbo& Fang, Alex Chengyu& Zhang, Xing. 2014. A Novel Feature Selection Strategy for Enhanced Biomedical Event Extraction Using the Turku System. BioMed Research International،Vol. 2014, no. 2014, pp.1-12.
https://search.emarefa.net/detail/BIM-454327
Modern Language Association (MLA)
Xia, Jingbo…[et al.]. A Novel Feature Selection Strategy for Enhanced Biomedical Event Extraction Using the Turku System. BioMed Research International No. 2014 (2014), pp.1-12.
https://search.emarefa.net/detail/BIM-454327
American Medical Association (AMA)
Xia, Jingbo& Fang, Alex Chengyu& Zhang, Xing. A Novel Feature Selection Strategy for Enhanced Biomedical Event Extraction Using the Turku System. BioMed Research International. 2014. Vol. 2014, no. 2014, pp.1-12.
https://search.emarefa.net/detail/BIM-454327
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-454327