Syawal, Asrul (2019) Sistem Pakar Deteksi Kenakalan Remaja Di Sekolah Menggunakan Metode Modified K-Nearest Neighbor (MKNN). Sarjana thesis, Universitas Brawijaya.
Abstract
Kenakalan remaja salah satu permasalahan yang dihadapi orang tua dalam membimbing anak. Kenakalan remaja sudah sering terjadi di dunia pendidikan teruitama disekolah hal ini menjadi perhatian bagi orang tua, selain orang tua, guru sebagai pembimbing di sekolah ikut aktif dalam membimbing siswa di sekolah agar terhindar dari kenakalan remaja dan berprilaku baik dan positif dilingkungan sekolah maupun lingkungan luar sekolah. Untuk menghindari kenakalan di sekolah dilakukan pencegahan dengan melakukan deteksi kenakalan yang mungkin akan terjadi pada anak berdasarkan faktor atau gejala-gejala perilaku yang sering dilakukan oleh anak. Pada penelitian ini dibuat suatu sistem untuk mendeteksi kenakalan remaja khsusnya disekolah untuk mencegah kenakalan yang mungkin terjadi menggunakan metode Modified K-Nearest Neighbor (MKNN). Metode MKNN merupakan metode perkembangan dari metode KNN, yang membedakan terdapat nilai validitas pada data latih untuk menghasilkan nilai yang lebih baik. Hasil dari pengujian ini menggunakan data latih sebanyak 60, dan akurasinya terhadap variasi data latih mendapatkan hasil akurasi ditingkat terendah ketika nilai variasi data latih sebesar 30% yaitu 76,78% dan akurasi ditingkat tertinggi ketika nilai variasi data latih sebesar 90% yaitu 100%. Dalam pengujian ini didapatkan rata-rata akurasi sistem maksimal sebesar 86,7%.
English Abstract
Juvenile delinquency is one of the problems faced by parents in guiding children. Juvenile delinquency has often happened in the world of primary education in schools, this is a concern for parents, besides parents, teachers as mentors in schools actively participate in guiding students in schools to avoid juvenile delinquency and to behave positively and positively in the school environment and outside the school environment . To avoid delinquency in school prevention is done by detecting delinquency that might occur in children based on factors or behavioral symptoms that are often carried out by children. In this study a system was created to detect juvenile delinquency especially in schools to prevent delinquency that might occur using the Modified K-Nearest Neighbor (MKNN) method. MKNN method is a developmental method of the KNN method, which distinguishes the value of validity in training data to produce better values. The results of this test used as much as 60 training data, and the accuracy of the variation in training data obtained the lowest level of accuracy when the value of training data variation was 30% the accuracy was 76.78% and the highest level of accuracy when the training data variation value was 90% the accuracy was 100%. In this test the average system accuracy was obtained at a maximum of 86.7%.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FILKOM/2019/327/051905645 |
| Uncontrolled Keywords: | Kenakalan Remaja, Kenakalan, Sistem Pakar, Modified K-Nearest Neighbor, K-Nearest Neighbor/Juvenile Delinquency, Mischief, Expert System, Modified K-Nearest Neighbor, K-Nearest Neighbor |
| Subjects: | 000 Computer science, information and general works > 005 Computer programming, programs, data > 005.4 Systems programming and programs > 005.43 Systems programs > 005.433 71 Systems programs (Education) |
| Divisions: | Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Endang Susworini |
| Date Deposited: | 24 Aug 2020 06:51 |
| Last Modified: | 24 Aug 2020 06:51 |
| URI: | http://repository.ub.ac.id/id/eprint/171705 |
Actions (login required)
 |
View Item |