Prediction and Diagnosis of Diabetic Retinopathy using Data Mining Technique

Hayrettin EVİRGEN; Menduh ÇERKEZİ

Makale detayları

Prediction and Diagnosis of Diabetic Retinopathy using Data Mining Technique

Yazarlar:

Hayrettin EVİRGEN | Menduh ÇERKEZİ

Yayıncı:

The association of science, education and technology

Dergi:

TOJSAT 2014 Vol.4 Issue 3, pp.32-37

Belge Tipi:

Makale

Yayın Tarihi:

7/2016

Belge Dili:

Türkçe

Document ID (Org No.):

2000835905

Sıklık:

Üç Aylık

pISSN:

2146-7390

Özet:

 

Diabetic retinopathy is the most common form of eye problem affecting people with diabetes, usually only affects people who have had diabetes for a long time period and can result in blindness. The aim of this study is to examine the naive Bayes algorithm which is one of the classification methods in data mining, and to analyze real life dataset in order to built predictive system for diabetic retinopathy disease. A total of 385 diabetes patients’ data were used to train the prediction system. All the categorical features in the dataset were selected by doctors and evaluation was made based on these features. The dataset was obtained at the Eye Clinic of the Sakarya University Educational and Research Hospital. It has been proven with cross-validation that naive Bayes algorithm can be used for diabetic retinopathy prediction with an improved accuracy of 89%.

Makale detayları

Prediction and Diagnosis of Diabetic Retinopathy using Data Mining Technique

Yazarlar: Hayrettin EVİRGEN | Menduh ÇERKEZİ

Yayıncı: The association of science, education and technology

Dergi: TOJSAT 2014 Vol.4 Issue 3, pp.32-37

Belge Tipi: Makale

Yayın Tarihi: 7/2016

Belge Dili: Türkçe

Document ID (Org No.): 2000835905

Sıklık: Üç Aylık

pISSN: 2146-7390

Sayı

Yazarlar:

Hayrettin EVİRGEN | Menduh ÇERKEZİ

Yayıncı:

The association of science, education and technology

Dergi:

TOJSAT 2014 Vol.4 Issue 3, pp.32-37

Belge Tipi:

Makale

Yayın Tarihi:

7/2016

Belge Dili:

Türkçe

Document ID (Org No.):

2000835905

Sıklık:

Üç Aylık

pISSN:

2146-7390