AURUM Mühendislik Sistemleri ve Mimarlık Dergisi
Yazarlar: Mariya KİKNADZE, Ahmet GÜRHANLI
Konular:Bilgisayar Bilimleri, Yapay Zeka
Anahtar Kelimeler:Artificial Neural Networks,Breast Cancer,Breast Cancer Diagnosis
Özet: Breast cancer is one of the most important malignant diseases in the world. In the United States, breast cancer ranks first among all oncological diseases in women and is the second leading cause of cancer mortality after lung cancer. Despite recent great success in the early detection and treatment of breast cancer, new approaches and algorithms are still being developed for early diagnosis. Breast cancer has many classifications, like other malignant diseases: histological, molecular, functional, TNM classification. Most cases of cancer can be diagnosed in the later stages of the disease, and treatment is often not responding and the patient is lost. Therefore, early detection of breast cancer is vital. This study uses the UCI Breast Cancer Wisconsin (Diagnostic) Data Set (WDBC), which is presented by measuring test classification accuracy, sensitivity, and specificity values. The data set was divided into 70% for the training phase and 30% for the testing phase. This study demonstrates the importance of optimization algorithm selectiona and parameters in the diagnosis of Breast Cancer using Artificial Neural Networks and investigates how they should be chosen. The accuracy results of different optimization algorithms and parameter values are reported.