Simple and accurate decision tree based on fuzzy stop criteria approach

Classification is an important task in data mining and machine learning while the decision tree is treated as one of the main algorithms considered in this area. Although decision tree make a comprehensible model but suffer disadvantage of complexity. In this paper, we proposed a novel decision tree based on fuzzy stop criteria (DTFSC) with the aim of simplifying tree and retaining their accuracy. For this reason, we used depth and standard error to achieve tradeoff between complexity and accuracy. Experimental results show that DTFSC outperforms its traditional counterpart (C4.5) in term of accuracy and complexity.