E-Handrawn Calculator

• Main Author: Mohamad, Syamimi
• Format: Final Year Project
• Language: English
• Published: Universiti Teknologi Petronas 2008
• Subjects: T Technology (General)
• Online Access: http://utpedia.utp.edu.my/7161/1/2008%20-%20E%20-handrawn%20calculator.pdf; http://utpedia.utp.edu.my/7161/

This draft for final report is about my research progress in developing e-Hand-Drawn Calculator as final year project. The purpose of this project is to demonstrate an application of back-propagation network (comparison of training their algorithms and transfer function) in order to developing e-Hand-Drawn Calculator. Back-propagation network is a supervised learning method, and is an implementation of the Delta rule. It requires a teacher that knows, or can calculate, the desired output for any given input. It is most useful for feed-forward networks (networks that have no feedback, or simply, that have no connections that loop). The term is an abbreviation for "backwards propagation of errors". Backpropagation requires that the activation function used by the artificial neurons (or "nodes") is differentiable. The main activities in this project are Assemble the training data, Create the network object, Train the network and Simulate the network response to new inputs. The training data consist of tern sample of each number zeros until number nine and symbol plus, minus, division and multiplication. These all data will be train, testing and validation before enter to next stage which is creating network object. This section presents the architecture of the network that is most commonly used with the backpropagation .algorithm; the multilayer feedforward network. The investigation for combination of Neuron Model (tansig, logsig, purelin) and training algorithms (traingd, traingdm, traingda, traingdx, trainrp, traincgp, traincgb, trainscg, trainbfg, trainoss, trainlm, trainbr); tend to know which combination will give the greatest result and smallest error.