Conventional classifiers such as k-nearest neighbor and support vector machine have been widely used,but in the big data era,more training will greatly reduce the training efficiency and accuracy of the classifier.To solve this problem,the pro-totype optimization method is used to filter and compress the existing training data to filter out a large number of redundant da-ta.The compressed data set is used as the prototype to train the classifier to improve the training efficiency and classification accuracy.The existing methods,designs a new classifier based on prototype optimization method,which can greatly reduce the training amount and ensure the classification accuracy.The test results verify the effectiveness of the proposed method.
维普
万方数据