Bagging neural network sensitivity analysis for feature reduction for in-silico drug design

by: MJ Embrechts, F Arciniegas, M Ozdemir, CM Breneman, K Bennett, L Lockwood

International Joint Conference on Neural Networks (IJCNN), Vol. 4 (2001), pp. 2478-2482 vol.4.

View FullText article

DOI, IEEE Explore

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

sona

Find related articles with these CiteULike tags

dimensionality-reduction, embedded-feature-selection, feature-selection, neural-networks, sensitivity-analysis

Abstract

This paper illustrates a new approach to sensitivity analysis for feature selection using multiple ensemble neural networks in a bootstrapping mode with bagging. This methodology is applied to in-silico drug design with QSAR (quantitative structural activity relationship), which is notoriously challenging for machine learning because typically there are on the order of 300-1000 dependent features, often for as few as 50-100 data points. For an HIV dataset with 160 wavelets descriptors, the number of relevant features was reduced to 35, and the resulting predictive neural network model gave better results than with the full feature set

@inproceedings{embrechts01bagging, abstract = {This paper illustrates a new approach to sensitivity analysis for feature selection using multiple ensemble neural networks in a bootstrapping mode with bagging. This methodology is applied to in-silico drug design with QSAR (quantitative structural activity relationship), which is notoriously challenging for machine learning because typically there are on the order of 300-1000 dependent features, often for as few as 50-100 data points. For an HIV dataset with 160 wavelets descriptors, the number of relevant features was reduced to 35, and the resulting predictive neural network model gave better results than with the full feature set}, author = {Embrechts, M. J. and Arciniegas, F. and Ozdemir, M. and Breneman, C. M. and Bennett, K. and Lockwood, L. }, booktitle = {Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on}, citeulike-article-id = {2646542}, doi = {10.1109/IJCNN.2001.938756}, journal = {International Joint Conference on Neural Networks (IJCNN)}, keywords = {dimensionality-reduction, embedded-feature-selection, feature-selection, neural-networks, sensitivity-analysis}, pages = {2478--2482 vol.4}, posted-at = {2008-04-09 18:02:12}, priority = {3}, title = {Bagging neural network sensitivity analysis for feature reduction for in-silico drug design}, url = {http://dx.doi.org/10.1109/IJCNN.2001.938756}, volume = {4}, year = {2001} }

RIS record

TY - CONF ID - embrechts01bagging L3 - citeulike-article-id:2646542 TI - Bagging neural network sensitivity analysis for feature reduction for in-silico drug design T2 - Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on JF - International Joint Conference on Neural Networks (IJCNN) VL - 4 SP - 2478 EP - 2482 vol.4 N2 - This paper illustrates a new approach to sensitivity analysis for feature selection using multiple ensemble neural networks in a bootstrapping mode with bagging. This methodology is applied to in-silico drug design with QSAR (quantitative structural activity relationship), which is notoriously challenging for machine learning because typically there are on the order of 300-1000 dependent features, often for as few as 50-100 data points. For an HIV dataset with 160 wavelets descriptors, the number of relevant features was reduced to 35, and the resulting predictive neural network model gave better results than with the full feature set KW - dimensionality-reduction KW - embedded-feature-selection KW - feature-selection KW - neural-networks KW - sensitivity-analysis AU - Embrechts, MJ AU - Arciniegas, F AU - Ozdemir, M AU - Breneman, CM AU - Bennett, K AU - Lockwood, L PY - 2001/// UR - http://dx.doi.org/10.1109/IJCNN.2001.938756 ER -

RIS BibTeX