Multi-class Protein Sequence Classification Using Fuzzy ARTMAP Export

@proceedings{mohammed2006, abstract = {The classification of protein sequences into families is an important tool in the annotation of structural and functional properties to newly discovered proteins. We present a classification system using pattern recognition techniques to create a numerical vector representation of a protein sequence and then classify the sequence into a number of given families. We introduce the use of fuzzy ARTMAP classifiers and show that coupled with the genetic algorithm based feature subset selection, the system is able to classify protein sequences with an accuracy of 93\%. This accuracy is compared with numerous other classification tools and demonstrates that the fuzzy ARTMAP is suitable due to its high accuracy, quick training times and ability for incremental learning.}, author = {Mohamed, S. and Rubin, D. and Marwala, T.}, booktitle = {Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on}, citeulike-article-id = {5405202}, citeulike-linkout-0 = {http://dx.doi.org/10.1109/ICSMC.2006.384960}, citeulike-linkout-1 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4274094}, comment = {SCOP/ASTRAL sequence classification Comparison to other techniques (GLM, RBF, MLP, k-Nearest Neighbor) References * Review on sequence alignment techniques * Review on alignment-free sequence comparison * discussion on difference between local and global features}, doi = {10.1109/ICSMC.2006.384960}, journal = {Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on}, keywords = {artmap, bioinformatics, classification, fuzzy\_artmap, fuzzy\_logic, protein\_sequence\_classification}, pages = {1676--1681}, posted-at = {2009-08-10 15:47:28}, priority = {0}, title = {Multi-class Protein Sequence Classification Using Fuzzy ARTMAP}, url = {http://dx.doi.org/10.1109/ICSMC.2006.384960}, volume = {2}, year = {2006} }

TY - CONF ID - mohammed2006 L3 - citeulike-article-id:5405202 N2 - The classification of protein sequences into families is an important tool in the annotation of structural and functional properties to newly discovered proteins. We present a classification system using pattern recognition techniques to create a numerical vector representation of a protein sequence and then classify the sequence into a number of given families. We introduce the use of fuzzy ARTMAP classifiers and show that coupled with the genetic algorithm based feature subset selection, the system is able to classify protein sequences with an accuracy of 93%. This accuracy is compared with numerous other classification tools and demonstrates that the fuzzy ARTMAP is suitable due to its high accuracy, quick training times and ability for incremental learning. JF - Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on EP - 1681 TI - Multi-class Protein Sequence Classification Using Fuzzy ARTMAP VL - 2 T2 - Systems, Man and Cybernetics, 2006. SMC '06. IEEE International Conference on SP - 1676 KW - artmap KW - bioinformatics KW - classification KW - fuzzy_artmap KW - fuzzy_logic KW - protein_sequence_classification N1 - SCOP/ASTRAL sequence classification Comparison to other techniques (GLM, RBF, MLP, k-Nearest Neighbor) References * Review on sequence alignment techniques * Review on alignment-free sequence comparison * discussion on difference between local and global features AU - Mohamed, S AU - Rubin, D AU - Marwala, T PY - 2006/// UR - http://dx.doi.org/10.1109/ICSMC.2006.384960 ER -