Abstract：KNN algorithm is a classification algorithm that is simple and easy to implement， but when the training set is rather big and features are more， its efficiency is low with which takes more time. To solve this problem， an improved KNN classification algorithm was proposed based on Fuzzy C-Means. The improved algorithm introduces the theory of Fuzzy C-Means based on the traditional KNN classification algorithm. Through processing the sample data clustering， the formation of sub clusters substitutes all the sample set of the sub cluster， which helps reduce the number of training set. Thereby the workload of the KNN classification process is reduced， with the classification efficiency improved and the KNN algorithm better applied in data mining. The theoretical analysis and experimental results show that this method is able to significantly improve the efficiency and accuracy of the algorithm when dealing with large data， meeting the demand of processing data.
2017, Vol. 51 
