Intelligent selection system for teachers’ replacement using feed forward neuralnetwork / Che Nor Syairah Che Rozaid

Prediction technique for teachers’ replacement has been reported to make better selection of candidates that meet all criteria as a teacher. The prediction is made based on their personal information that they fill in the application form. Before this, process to make selection is done manually and more applicants complained that the reaction of their application is too late and so slow. Therefore this intelligent teacher’s selection prototype has been developed to reduce the time taken to response each application and also to make better selection.
There are many techniques that can be used to make prediction process but the technique that is selected to use for this system is Neural Network. Neural Network has two types that can be used such as feed forward and backward technique.
The prediction is done using Neural Network technique where some of applicant’s information is taken to predict to see they are eligible or not to be a teacher. The first purpose this prototype is create to make prediction of some data to measure the accuracy rate either same or better than manually process before this.
The result that are get after using this system is the accuracy of selection teacher’s replacement can be improved. The percentages of actual output same with the desired output is more than 80%. In the Neural Network training, testing data is repeatedly done for the purpose to get the better weight value where it will be used to predict the future data.