![Open Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextgreen.png)
![Restricted Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextred.png)
![Open Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextgreen.png)
![Open Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltext_open_medium.gif)
![Restricted Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltextred.png)
![Restricted Access](https://i-scholar.in/lib/pkp/templates/images/icons/fulltext_restricted_medium.gif)
Radial Basis Function Network Construction Using Modified Gram Schmidt Algorithm
Subscribe/Renew Journal
The Neural network provides the frame work of self-learning system. It allows the user to design the work around huge amount of data, in dynamic and nonlinear form. By using this system, experts, technical peoples and developers in this domain are trying to make the widely used system in form of autonomous system. The Radial Basis Function Neural Network is come under the supervised type learning neural network. It has been widely used in neural network problems like classification, signal processing, pattern recognition, fault diagnosis etc. The construction of any type of neural network is very an important task. Network must be constructed in such way that it can adapt to any of problem and learn the network with computational efficient and compact manner. While building initial model of Radial Basis Function Network important issues are to determine of hidden layer parameters (center and width) and output weights. A proposed algorithm for Radial Basis Function Neural Network construction is suggested to deal with important issues of Network construction. It is designed for classification problem. A suggested algorithm is mainly divided into four parts that are K-mean algorithm, initial model construction, and performance evaluation and based on MSE value initially computed output weights are optimized using Particle Swarm Optimization method. Again the optimized Radial Basis Function network is constructed based new value of output weights. Final, part is performance evaluation of optimized model and initial model. To do the analysis of both models it is trained and tested on datasets from UCI repositories. Analysis of both networks is carried out on testing samples. The performance of training, testing phases of network are measured based upon classification accuracy, number of samples misclassified and time required for computation.
Keywords
![](https://i-scholar.in/public/site/images/abstractview.png)
Abstract Views: 188
![](https://i-scholar.in/public/site/images/pdfview.png)
PDF Views: 3