Document Detail


Feature selection for the imbalanced QSAR problems by using easyensemble.
MedLine Citation:
PMID:  20063462     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
Activities of drug molecules can be predicted by Quantitative Structure Activity Relationship (QSAR) models, which overcomes the disadvantages of high cost and long cycle by employing the traditional experimental method. With the fact that the number of drug molecules with positive activity is rather fewer than that of negatives, it is important to predict molecular activities considering such an imbalanced situation. Here we propose one embedded feature selection algorithm i.e., Prediction Risk based feature selection for EasyEnsemble (PREE) to treat this problem and improve generalisation performance of the EasyEnsemble classifier. Experimental results on the drug molecules data sets show that PREE obtains better performance, compared with the asymmetric bagging and EasyEnsemble.
Authors:
Tian-Yu Liu; Guo-Zheng Li; Jack Y Yang; Mary Qu Yang
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't    
Journal Detail:
Title:  International journal of computational biology and drug design     Volume:  1     ISSN:  1756-0756     ISO Abbreviation:  Int J Comput Biol Drug Des     Publication Date:  2008  
Date Detail:
Created Date:  2010-01-11     Completed Date:  2010-03-24     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101479540     Medline TA:  Int J Comput Biol Drug Des     Country:  England    
Other Details:
Languages:  eng     Pagination:  334-46     Citation Subset:  IM    
Affiliation:
School of Electric, Shanghai Dianji University, Shanghai, China. liuty@sdju.edu.cn
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Algorithms
Drug Design*
False Positive Reactions
Humans
Models, Biological
Quantitative Structure-Activity Relationship*
Risk Assessment
Sensitivity and Specificity
Treatment Outcome

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  2009 and beyond: the decade of personalised medicine.
Next Document:  Pairwise statistical significance and empirical determination of effective gap opening penalties for...