Recognizing Handwritten Digits by Multilayer Perceptron along with Backpropagation Learning
ArunaLiyanaarachchi, Moayed D. Daneshyari

In this study, we have developed a machine learning algorithm to automatically recognize the handwritten digits. We have used a multilayer perceptron neural network and trained its weight using the backpropagation learning algorithm. Using the handwritten digits from the MNIST database, we created two sets of data, one for training the proposed network and another to test the network. The performance of the network has been tested with using different activation functions and overall results had been discussed on performance of the network on recognizing the handwritten digits.

Full Text: PDF     DOI: 10.15640/jcsit.v9n2a3