CIVILICA We Respect the Science
(ناشر تخصصی کنفرانسهای کشور / شماره مجوز انتشارات از وزارت فرهنگ و ارشاد اسلامی: ۸۹۷۱)

Diagnosis of Breast Cancer Using the Gradient Boosted Trees Data Mining Technique

عنوان مقاله: Diagnosis of Breast Cancer Using the Gradient Boosted Trees Data Mining Technique
شناسه ملی مقاله: SISOC01_048
منتشر شده در کنگره بین المللی جراحی سرطان شیراز در سال 1397
مشخصات نویسندگان مقاله:

Raziyeh Sarhadi - Iran, Shiraz, Teachers Square - North Iman - Sepah Bank - Apadana Institute of Higher Education
Reza Akbari - Iran, Shiraz, Blvd Masters, Industrial School
Sedigheh Tahmasebi - Breast Diseases Research Center, Shiraz University of Medical Sciences, Shiraz, Iran
Vahid Zangouri - Breast Diseases Research Center, Shiraz University of Medical Sciences, Shiraz, Iran

خلاصه مقاله:
Breast cancer is one of the deadliest and most common cancers among women in the world today. Data mining is one of the strongest and best practices in the field of diagnosis of breast cancer that has come to the aid of doctors in this area. This research was performed by working on randomized indigenous data in Shiraz, Iran, and modeling using three svm, regression linear, gradient boosted tree algorithms and also using ten fold cross validation method to verify the accuracy of the model. The results showed that gradient algorithm boosted trees with a precision of 99/74 showed a flipping performance over the rest of the algorithms.

کلمات کلیدی:
Breast cancer, Data mining, Ten fold cross validation, Gradient boosted Trees

صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/838447/