Conclusion

In this article, the problem of learning models or classifiers in a non-stationary environment has been studied. Suitable methods and techniques for learning effective classifiers for this type of environment have been presented and analyzed. These methods have been classified as shown in figure 11 . Each of these methods has its advantages and disadvantages. The choice of a method depends overall on the characteristics of the potential changes in the application, as shown in figure 12 .

There is very little work