A novel over-sampling method and its application to miRNA predictionReportar como inadecuado

A novel over-sampling method and its application to miRNA prediction - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

MicroRNAs miRNAs are short~22nt non-coding RNAs that play an indispensable role in gene regulation ofmany biological processes. Most of current computational, comparative, andnon-comparative methods commonly classifyhuman precursor micro- RNA pre-miRNA hairpins from both genome pseudohairpins and other non-coding RNAs ncRNAs. Although there were a fewapproaches achieving promising results in applying class imbalance learningmethods, this issue has still not solved completely and successfully yet by theexisting methods because of imbalanced class distribution in the datasets. Forexample, SMOTE is a famous and general over-sampling method addressing thisproblem, however in some cases it cannot improve or sometimes reduces classification performance. Therefore,we developed a novel over-sampling method named incre-mental- SMOTE todistinguish human pre-miRNA hairpins from both genome pseudo hairpins and otherncRNAs. Experimental results on pre-miRNA datasets from Batuwita et al. showed that our method achievedbetter Sensitivity and G-mean than the control no over- sampling, SMOTE,and several successsors of modified SMOTEincluding safe-level-SMOTE and border-line-SMOTE. In addition, we alsoapplied the novel method to five imbalanced benchmark datasets from UCI MachineLearning Repository and achieved improvements in Sensitivity and G-mean.These results suggest that our method outperforms SMOTE and several successorsof it in various biomedical classification problems including miRNA classification.


Imbalanced Dataset; Over-Sampling; SMOTE; miRNA Classification

Cite this paper

Dang, X. , Hirose, O. , Saethang, T. , Tran, V. , Nguyen, L. , Le, T. , Kubo, M. , Yamada, Y. and Satou, K. 2013 A novel over-sampling method and its application to miRNA prediction. Journal of Biomedical Science and Engineering, 6, 236-248. doi: 10.4236-jbise.2013.62A029.

Autor: Xuan Tho Dang, Osamu Hirose, Thammakorn Saethang, Vu Anh Tran, Lan Anh T. Nguyen, Tu Kien T. Le, Mamoru Kubo, Yoichi Yamada, Kenj

Fuente: http://www.scirp.org/


Documentos relacionados