Performance Analysis of K-Nearest Neighbor (KNN) Classifier for Optimum K Value

The K-nearest neighbor (KNN) is one of the simplest and accurate techniques for classification. K-nearest neighbors algorithm is an example of supervised learning. K-nearest neighbor can be used for predictive of classification and regression problems. In K-nearest neighbor, nearest neighbor is measured with respect to value of k. The value of K defines how many nearest neighbors are need to examine. The main advantages of K-nearest neighbors technique is its effectiveness for large training data. It is also robust to noisy training data. K-nearest neighbors is a lazy learning algorithm. In K-nearest neighbors there is no assumption for underlying data distribution. There are several algorithms and methods have been developed to solve the problem of classification. But for finding new algorithm is a process for improving accuracy and efficiency. In this paper we try to determine the best value for K, by continues trying a few values before settling final values.