Deteksi Diabetik Retinopati menggunakan Regresi Logistik

Autor: Raras Tyasnurita, Adhi Yoga Muris Pamungkas
Jazyk: English<br />Indonesian
Rok vydání: 2020
Předmět:
Zdroj: Ilkom Jurnal Ilmiah, Vol 12, Iss 2, Pp 130-135 (2020)
Druh dokumentu: article
ISSN: 2087-1716
2548-7779
DOI: 10.33096/ilkom.v12i2.578.130-135
Popis: Retinopathy diabetic is a disease caused by diabetes mellitus complications that can cause damage to the retina of the eye. It has a direct impact on the disruption of the vision of the patient. Detecting this disease is very important to prevent total blindness on diabetes mellitus patients. One method to do the detection is by using machine learning. This research uses feature extraction data from an image that contains signs of retinopathy diabetic or not. In this research, we focus on retinopathy diabetic classification. We applied logistic regression algorithm for classification. There is four training condition in a machine learning model: using the default parameter, feature standardization, feature selection, and hyper-parameter tuning. The model used a regularization control (C) value of 11.288, iterations 200, and a regularization penalty (l1). The experimental results show that this proposed model with full features produced 80,17% accuracy in data validation.
Databáze: Directory of Open Access Journals