A new classification algorithm WKS based on weight
Min Zhang, Min Qi, Kang Sun, Yujun Niu, Longxiang Shi · 2017
By analyzing the disadvantages of the traditional KNN using lazy learning that directly classify the data based on the K neighboring classes using the majority voting method, a new Sigmoid weighted classification algorithm WKS (Weighted KNN Based On Sigmoid) was proposed. WKS provides a new method for learning and training, since each training data diϵ D contributes to the correct classification of the classifier-expressed by weight Wi. WKS combined with the idea of AdaBoost algorithm that change the data weight distribution in the training base classifier, using Sigmoid function to achieve the weight update, the neighbors with correct contribution to the classify increasing the sample weight, otherwise decrease the weight. Firstly, The training set was modeled by training, and then use the testing set to do classification. By UCI data sets experiment, the experimental results show that WKS has better classification accuracy than traditional KNN algorithm.