TY - GEN
T1 - A projection based learning meta-cognitive RBF network classifier for effective diagnosis of Parkinson's disease
AU - Sateesh Babu, G.
AU - Suresh, S.
AU - Uma Sangumathi, K.
AU - Kim, H. J.
PY - 2012
Y1 - 2012
N2 - In this paper, we proposed a 'Projection Based Learning for Meta-cognitive Radial Basis Function Network (PBL-McRBFN)' classifier for effective diagnosis of Parkinson's disease. McRBFN is inspired by human meta-cognitive learning principles. McRBFN uses the estimated class label, the maximum hinge error and class-wise significance to address the self-regulating principles of what-to-learn, when-to-learn and how-to-learn in a meta-cognitive framework. Initially, McRBFN begins with zero hidden neurons and adds required number of neurons to approximate the decision surface. When a neuron is added, network parameters are initialized based on the sample overlapping conditions. The output weights are updated using a PBL algorithm such that the network finds the minimum point of an energy function defined by the hinge-loss error. The experimental results on parkinson's data sets based on vocal and gait features clearly highlight the superior performance of PBL-McRBFN classifier over results reported in the literature for detection of individual with or without PD.
AB - In this paper, we proposed a 'Projection Based Learning for Meta-cognitive Radial Basis Function Network (PBL-McRBFN)' classifier for effective diagnosis of Parkinson's disease. McRBFN is inspired by human meta-cognitive learning principles. McRBFN uses the estimated class label, the maximum hinge error and class-wise significance to address the self-regulating principles of what-to-learn, when-to-learn and how-to-learn in a meta-cognitive framework. Initially, McRBFN begins with zero hidden neurons and adds required number of neurons to approximate the decision surface. When a neuron is added, network parameters are initialized based on the sample overlapping conditions. The output weights are updated using a PBL algorithm such that the network finds the minimum point of an energy function defined by the hinge-loss error. The experimental results on parkinson's data sets based on vocal and gait features clearly highlight the superior performance of PBL-McRBFN classifier over results reported in the literature for detection of individual with or without PD.
UR - http://www.scopus.com/inward/record.url?scp=84865109404&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84865109404&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-31362-2_67
DO - 10.1007/978-3-642-31362-2_67
M3 - Conference contribution
AN - SCOPUS:84865109404
SN - 9783642313615
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 611
EP - 620
BT - Advances in Neural Networks, ISNN 2012 - 9th International Symposium on Neural Networks, Proceedings
T2 - 9th International Symposium on Neural Networks, ISNN 2012
Y2 - 11 July 2012 through 14 July 2012
ER -
