Fig. 3From: Implementation of ensemble machine learning algorithms on exome datasets for predicting early diagnosis of cancersDistortion (error rate) versus number of clusters (K-value) generated by elbow curve method to determine the best K-value for KNN. From the graph, we can see that the error rate drastically drops around 14, thereafter the decrease in error rate is low. Therefore, the ideal cluster value should be between 11 and 13Back to article page