A method to select a good setting for the kNN algorithm when using it for breast cancer prognosis

Breast cancer is the world's second most frequent type of cancer and in Japan it is the third most frequent one. The prognosis of its recurrence, after a first treatment, is very important to increase the survival rate of a patient. This work shows the application of the k-Nearest Neighbors (kNN) method to prognosis breast cancer and also proposes a method to select a good setting with the parameters that can be changed when using this classification method. Using our method with the Wisconsin's breast cancer prognosis data, the kNN method has an average accuracy of 76%, a small standard deviation, and a small difference between its maximum and minimum values.