Concept drift detection method based data stream classification model
Nian X. Sun · Jisuanji gongcheng yu sheji · 2013
To overcome the effect of the data stream concept drift phenomenon for data stream classification,improve the accuracy of classification,a concept drift detection method based data stream classification model is presented.Firstly,the related information of concept drift is introduced.For different concept drift types,there are different kinds of concept drift detection methods for recognizing.Using the proposed concept drift detection method,the updating process of the data stream classification model can be controlled in reason,and the accuracy of classification can also be improved.The experiment employs two kinds of data stream for testing the validation and correctness of the proposed model.In addition,some traditional classification models are also used for comparing test.