Open AccessPrediction of courses score using Artificial Neural Network with Backpropagation algorithm
Author(s) -
Dede Kurniadi,
Asri Mulyani,
Yosep Septiana,
Ismail Yusuf
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1098/3/032110
Subject(s) - backpropagation , artificial neural network , computer science , anticipation (artificial intelligence) , multilayer perceptron , perceptron , mean squared error , attendance , algorithm , machine learning , artificial intelligence , statistics , mathematics , economics , economic growth
This study aims to predict the final score of courses of students who are undergoing education in higher education. In the future, this work will be used to inform students that there must be early anticipation in implementing the study of courses in order to get a satisfying final grade. The Artificial Neural Network with Backpropagation algorithm is used to solve the prediction problem for the final grade of this course. The dataset collected was 337 students who will be used for training and testing needs. Prediction variables used consisted of the value of attendance, assignments, midterms, and final exams. The network architecture using a multi-layer perceptron with three layers, namely the input layer, hidden layer, and output layers, with a 4-3-1 neuron pattern. The results showed that using a learning rate of 0.2 and repetition Epoch 1000 times, resulting in an RMSE value of 0.040929 and MSE of 0.001675 with an accuracy rate of 93.43%.