Eko Apriansyah Saputra, - (2021) PREDIKSI ATTETION DEFICIT HYPERACTIVITY DISORDER MENGGUNAKAN JARINGAN SYARAF TIRUAN BACKPROPAGATION MOMENTUM. Skripsi thesis, Universitas Islam Negeri Sultan Syarif Kasim Riau.

Preview	Text KECUALI BAB IV & V.pdf Download (5MB) | Preview
Preview	Text BAB IV & V.pdf Download (7MB) | Preview

Abstract

One common thought that is still developing in society is when a child has problems in the learning process because the child is difficult to focus, difficult to organize or likes to walk around in the classroom, parents will scold the child and conclude that the child is stupid. In the world of psychology, this is a behavioral disorder called Attention Deficit Hyperactivity Disorder or ADHD. If ADHD is not treated immediately, the most severe effect it will cause is the threat of long-term intelligence. In determining a child with ADHD, a psychologist must carry out a diagnosis by checking the symptoms according to the DSM V book so that it will take time. To make it easier for psychologists to work faster, an application that can detect ADHD is needed. The method used is Backpropagation Momentum with 19 variables input and 3 ADHD outputs, namely class 1 (predominant incentive), class 2 (predominantly hyperactive-impulsive), class 3 (combination). The parameters used were the learning rate 0.01, 0.1, and 0.2 and the momentum parameters used were 0.25, 0.4 and 0.75, the maximum initialization epoch = 1000, learning rate (α) = 0.2, Momentum (µ) = 0.75, target error 0.001. The data comparisons used were 70:30, 80:20, and 90:10. The best accuracy is obtained with a learning rate parameter of 0.2 on all momentum parameters in the 90:10 data section using 124 training data and 14 test data resulting in 100% accuracy. Thus the Backpropagation Momentum method can be applied for ADHD detection.

Actions (login required)

View Item