Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/160326
Title: Perbandingan Kinerja Model Logistic Regression dan Artificial Neural Network dalam Mendiagnosis Pasien Diabetes
Other Titles: 
Authors: Julianto, Mochamad Tito
Mangku, I Wayan
Noviyanti, Desi
Issue Date: 2024
Publisher: IPB University
Abstract: Diabetes melitus merupakan kondisi yang terjadi akibat kekurangan produksi insulin, baik sebagian maupun keseluruhan, yang menyebabkan peningkatan kadar glukosa dalam darah. Mendeteksi diabetes sejak dini sangat penting untuk mencegah berkembangnya komplikasi yang serius. Penelitian ini bertujuan untuk membandingan kinerja model Logistic Regression dan Artificial Neural Network(ANN) dalam mendiagnosis pasien diabetes. Data penelitian terdiri dari 8 variabel bebas dan 1 variabel target. Hasil penelitian menunjukkan bahwa model ANN merupakan model terbaik berdasarkan hasil metrik evaluasi akurasi dan nilai AUC, baik pada data training (akurasi = 81.82% dan AUC = 89.53%) maupun data testing (akurasi = 80.73% dan AUC = 87.53%). Sementara itu, model Logistic Regression merupakan model terbaik berdasarkan kecepatan waktu komputasinya, baik pada data training (0.0010 detik) maupun data testing (0.0003 detik).
Diabetes mellitus is a metabolic disorder characterized by insufficient insulin production, leading to elevated glucose levels in the bloodstream. Early detection of diabetes is crucial to prevent the development of serious complications. This study aims to compare the performance of Logistic Regression and Artificial Neural Network (ANN) models in diagnosing diabetes patients. The dataset consists of eight independent variables and one target variable. The findings reveal that the ANN model outperforms the Logistic Regression model based on accuracy and AUC (Area Under Curve) metrics. The ANN model achieved an accuracy of 81.82% and an AUC of 89.53% on the training data, and an accuracy of 80.73% and an AUC of 87.53% on the testing data. Conversely, the Logistic Regression model excelled in computational speed, with processing times of 0.0010 seconds for the training data and 0.0003 seconds for the testing data.
URI: http://repository.ipb.ac.id/handle/123456789/160326
Appears in Collections:UT - Mathematics

Files in This Item:
File Description SizeFormat 
cover_G5401201022_24dd8ac036514e2293d98c99b3d21a6d.pdfCover296.19 kBAdobe PDFView/Open
fulltext_G5401201022_48eafeea81c549e893cfbcb9c0b3ddab.pdf
  Restricted Access
Fulltext1.4 MBAdobe PDFView/Open
lampiran_G5401201022_4e0e2640ed334fc288897e0e98991264.pdf
  Restricted Access
Lampiran202.37 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.