Neural Network Classification for Breast Cancer Analysis
Paquita Putri Ramadhani , Intan Nurma Yulita
Department of Computer Science, Padjadjaran
University, Sumedang 45363,Indonesia
Abstract
Breast cancer is a type of cancer that has a high mortality rate. Cancer is caused by a lump from a collection of cells that grows and attacks the surrounding tissue. Most breast lumps are benign, but the benign breast lumps can increase the risk of developing breast cancer. It is important for everyone to check for a lump in the breast. It is done to find out whether the lump is potentially cancer or not. Early detection can provide better handling. The diagnosis of breast cancer by a doctor is done by analyzing several factors. To help the doctor in diagnosing the data efficiently, this study implemented machine learning. The diagnosis was based on the features of digital image computation in the process of fine-needle aspiration (FNA) of a breast mass. The data came from 569 patients. The study used the neural networks classification method with multilayer perceptron algorithms. The results were obtained that the use of neural networks gave higher accuracy if it compared it in the ZeroR method. Their accuracies were 95.96%, and 62.74%, respectively.
Keywords: Breast cancer, Neural Network Classification, ZeroR dan Multilayer Perceptron
Topic: Computer Science