2026-08-24T10:27:30Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/3267 2021-12-13T00:27:08Z 1582963302567:1597824273898 user-ucsy

Comparison of Normal Neural Network Ensemble and Clustering Based Neural Network Ensemble Naing, Hnin Hnin Nyunt, Thi Thi Soe Artificial neural networks (ANNs) arecomputing models for information processingand pattern identification. An ANN is a networkof many simple computing units called neuronsor cells, which are highly interconnected andorganized in layers. Ensemble neural network isa learning paradigm where several neuralnetworks are jointly used to solve a problem.Generalization ability of a neural network can besignificantly improved through ensemblingneural networks, i.e. training several neuralnetworks and combining their results in someway. Ensemble neural network is a collection of a(finite) number of neural networks that aretrained for the same task. Since it behavesremarkably well and is easy to use, ensembleneural network is regarded as a promisingmethodology that can profit not only experts inneural computing but also ordinary engineers inreal world applications. This paper presents theensemble neural network method trained withclustering can improve the accuracy of theclassifier than single neural network. The systemis test with three datasets from UCI machinelearning repository and results are presented. 2017-12-27 http://hdl.handle.net/20.500.12678/0000003267 https://meral.edu.mm/records/3267