¹Institute of Scientific and Technical Information of China, Beijing, 100038, China
²Tianjin Key Lab of Intelligent Computing & Novel software Technology, Tianjin University of Technology, China
³Key Laboratory of Computer Vision and System (Tianjin University of Technology), Ministry of Education, China

---

Abstract—It’s well known machine learning from examples is an effective method to solve non-linear classification problem. A new dynamic method of machine learning from transition example is given in this paper. This method can improve the traditional method ID3 which learns from static eigenvalues of examples. The limits of the traditional method ID3 lie on no comprehension and no memory, especially, no the varieties and dynamic correlation of eigenvalues. In the new method, it can learn from dynamic eigenvalues, the change of data can be learned because the training data is the initial eigenvalue and the end eigenvalue in the interval. All eigenvalue’s varieties and correlation can be understood and remembered in application. By test experiments, the new method can be used as classifier when the multi-parameters are dynamic correlation, and it has special use in the many kinds of information fusion fields.

Index Terms—machine learning, entropy, dynamic, classifier, example

[PDF]

Cite: Xiao-dan Zhang, De-gan Zhang De-xin Zhao, Xue-Jing Kang, Xiao-dong Qiao, "A New Dynamic Method of Machine Learning From Transition Examples," Journal of Software vol. 6, no. 10, pp. 2064-2067, 2011.