SISTEM PAKAR DETEKSI DINI PENYAKIT STROKE MENGGUNAKAN METODE K-NEAREST NEIGHBOR

Anisa, Adelia Lailla (2022) SISTEM PAKAR DETEKSI DINI PENYAKIT STROKE MENGGUNAKAN METODE K-NEAREST NEIGHBOR. Skripsi thesis, UNIVERSITAS BHAYANGKARA SURABAYA.

[img] Text
1814311049_COVER_DALAM.pdf
Download (158kB)
[img] Text
1814311049_LEMBAR_PENGESAHAN.pdf
Download (765kB)
[img] Text
1814311049_LEMBAR_PERSETUJUAN.pdf
Download (501kB)
[img] Text
1814311049_BAB 1.pdf
Download (226kB)

Abstract

Stroke is an attack on the brain that can occur suddenly with the result of death or paralysis of one part of the body, this occurs because blood flow to the brain is cut off. Stroke is the third leading cause of death in developed countries after heart and cancer. To reduce the high mortality rate due to this disease prevention is carried out. This prevention can be done through a doctor's examination. The number of factors that can trigger stroke causes doctors to experience problems in carrying out early detection. These unstructured parameters can be solved using an expert system. An expert system is a system that is used to solve real-life problems. His knowledge comes from an expert. The expert system in this study uses the K-Nearest Neighbor (KNN) method in early detection. This study can detect the level of stroke risk from high, moderate, to low. The test results show that the manual calculation test with the system calculation test got an accuracy value of 100%, Meanwhile, for system testing with interview results with a value of K = 1, the level of conformity is 85%, K = 3 is 70%, K = 5 is 70%, and K = 10 is 65%. Training data used was 46 with 20 test data. For the research process, it begins with consulting with a stroke expert, dr Yenny Purwoyati Haerume. After conducting the consultation, the next step is to analyze the raw data which is then computed using the KNN method.

Item Type: Thesis (Skripsi)
Uncontrolled Keywords: K-Nearest Neighbor, KNN, Stroke, Expert System
Depositing User: Perpus Ubhara Surabaya
Date Deposited: 28 Feb 2023 04:13
Last Modified: 28 Feb 2023 04:13
URI: http://eprints.ubhara.ac.id/id/eprint/1747

Actions (login required)

View Item View Item
UBHARA Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.