Muhammad Mumtaz Ramadhan, . (2023) KLASIFIKASI PENYAKIT DIABETES MENGGUNAKAN METODE SUPPORT VEKTOR MACHINE (SVM). Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (118kB) |
|
Text
AWAL.pdf Download (473kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (146kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (771kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (286kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (2MB) |
|
Text
BAB 5.pdf Download (121kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (142kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (148kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (205kB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (17MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (864kB) |
Abstract
Diabetes is a quite dangerous disease. There are several indicators or factors that influence diabetes. For example, age, gender, weight, and so on. Health workers or doctors have difficulty determining what factors are more likely for someone to develop diabetes and which patients can be said to be positive or negative for diabetes. The aim of this research is to make it easier for doctors or health experts to make predictions, decisions, and/or classify patients who are positive for diabetes and negative. This research uses SVM and KNN methods. Then compared, which methods are better for predicting or classifying diabetes. Based on the results of the research that has been done, the accuracy value obtained with the SVM model is quite good, because the results of the evaluation metrics obtained results of more than 80% with the accuracy value obtained is 88% with cross val accuracy of 89%, precision of 83%, recall of 100%, F1 Score of 90%, and ROC obtained of 83%. While the best accuracy obtained with the KNN model is 99% with K1. Cross Val Accuracy obtained was 96%, precision was 100%, recall obtained was 98%, F1 was 99%, and ROC obtained was 83%. Thus, in this research, the KNN method is better than SVM in predicting or classifying diabetes.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 1910511102] [Pembimbing: Widya Cholil] [Penguji 1: Didit Widiyanto] [Penguji 2: Rio Wirawan] |
| Uncontrolled Keywords: | Diabetes, Classification, Prediction, SVM, KNN |
| Subjects: | Q Science > Q Science (General) R Medicine > RA Public aspects of medicine R Medicine > RT Nursing |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | Muhammad Mumtaz Ramadhan |
| Date Deposited: | 16 Feb 2024 08:45 |
| Last Modified: | 16 Feb 2024 08:45 |
| URI: | http://repository.upnvj.ac.id/id/eprint/27514 |
Actions (login required)
 |
View Item |