Eko Saputra, - (2021) PENERAPAN METODE K-NEAREST NEIGHBOR UNTUK KLASIFIKASI AUTISM SPECTRUM DISORDER. Skripsi thesis, Universitas Islam Negeri Sultan Syarif Kasim Riau.

Preview	Text Laporan BAB IV dan V.pdf Download (1MB) | Preview
Preview	Text Laporan Full Final.pdf Download (2MB) | Preview

Abstract

Autism Spectrum Disorder (ASD) is a behavioral and developmental disorder that causing social development, language skills, caring of the environment, living in their own world, emotional and intellectual disorders. Autism can also be caused by genetic factors. Psychologists diagnose ASD children by means of a test using a questionnaire, related to the symptoms the child has. Symptoms can refer to the DSM-IV or DSM-V books, which must always be opened sheet by sheet. This takes a long time in the detection process. Therefore, to help psychologists know that children have ASD or non-ASD quickly, an application is needed to classify ASD using the k-Nearest Neighbors (k-NN) method using 13 input variables and 1 output, namely ASD or non-ASD. K-NN is a supervised learning algorithm where the results of new instances are classified based on the majority of the k-nearest neighbor category to determine ASD and non-ASD classes. The results obtained from this study explained that the K-NN method was able to classify ASD and non-ASD children well. The highest accuracy obtained is 96.76% where the accuracy is at the value of k 3 with a data comparison of 90:10 on the fold to 1.

Actions (login required)

View Item