Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Muhammad Hafidzullah
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Seleksi Fitur dengan Information Gain pada Identifikasi Jenis Attention Deficit Hyperactivity Disorder Menggunakan Metode Modified K-Nearest Neighbor Muhammad Hafidzullah; Sutrisno Sutrisno; Marji Marji
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 11 (2019): November 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (363.05 KB)

Abstract

Attention Deficit Hyperactivity Disorder (ADHD) is a disorder that usually occurs in early childhood. In general there are three types of behavior associated with this disorder, namely: inattentiveness, impulsiveness, and hyperactivity. ADHD mental illness can only be recognized through changes in patient behavior. In this study, the data used using 45 features, where the number of features can affect the performance and accuracy of the classification method. Feature selection aims to reduce features that are of equal importance to make classification algorithms easier to operate more quickly and effectively so as to produce better accuracy. The method to be used for the selection of features is the Information Gain method. The use of the Information Gain method in this case is to select the best features that have relevance to the related data. These features are selected based on the magnitude of the gain value obtained, where the greater the gain value, the more relevant the feature is with the related data. The best average accuracy results obtained from this system are obtained in the second test scenario with the highest average accuracy value of 88% obtained at k = 37 and k = 42 and the number of features 36 and 41, and at k = 1 in testing without using Information Gain. These results indicate the use of the Information Gain feature selection method in this case has a fairly good accuracy value.
Co-Authors Marji Marji Sutrisno Sutrisno