Recognition of Off-line Handwritten Chinese Character by Using Decision Tree Based on Hiberarchy Decomposition

Decision tree based on hiberarchy decomposition is a kind of improved ID3 algorithm. It splits the training set by choosing different key attributes in different layers according to the correlation between classes and attributes. Compared with traditional ID3 algorithm, its rules are simpler and more general. This paper uses the hiberarchy decomposition methods as well as the C4.5 algorithm and makes some adjustments to deal with the recognition of off-line handwritten Chinese character by constructing a multi-level decision tree. At last, get a scheme of rough classification and analyze the results with different attributes. Compared with the single decision tree, the decision tree based on hiberarchy decomposition has more advantages when dealing with the multi-class problem. Experiment results show that this new method has better accuracy rate.