Nugroho, Syamsul (2020) PERBANDINGAN METODE FUZZY K-NEAREST NEIGHBOR DAN NEIGHBOR WEIGHTED K-NEAREST NEIGHBOR UNTUK DETEKSI PENYAKIT STROKE. Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
NASKAH_PUBLIKASI-Syamsul Aji Nugroho-5150411038.pdf Download (888kB) | Preview |
Abstract
Stroke or Cerebrovascular Accident (CVA) is a disorder of nerve function caused by disruption of blood flow in the brain and causes interference with functional activity. Stroke is the third most common cause of death in developed countries, after heart disease and cancer. With the continued increase in stroke patients each year, there has not been an effective effort in tackling the disease either by increasing public awareness or optimal management of stroke. Data Mining is a series of processes to explore the added value of a large data set in the form of knowledge that has not been known manually. K-Nearest Neighbor is the simplest method, easy to implement just by setting one parameter k. But K-Nearest Neighbor also has several major weaknesses. To overcome these deficiencies, one way to do that is to make improvements by estimating class probabilities. This study compares the method of Fuzzy K-Nearest Neighbor and Neighbor Weighted K-Nearest Neighbor to stroke detection. Test results with differences in the amount of test data that is 50, 70, 90, 150 and 200 and using a neighbor value (k) that is 17 to 21 and exponential value 2, then the average accuracy of 81,272% and 81,814% is obtained by using balanced data , while the unbalanced data are 82.45% and 82.75%. Keywords: Neighbor Weighted K-Nearest Neighbor, Fuzzy K-Nearest Neighbor, Classification
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 26 Mar 2020 01:06 |
| Last Modified: | 26 Mar 2020 01:06 |
| URI: | http://eprints.uty.ac.id/id/eprint/4849 |
Actions (login required)
 |
View Item |
