Classification Techniques of Datamining to Identify Class of the Text with Fuzzy Logic
Renuka D. Suryawanshi, D. M. Thakore · 2012
Decision tree (DT) is a very practical and popular approach in the machine learning domain for solving classification problems in data mining . Decision tree learning algorithm has been successfully used in expert systems in capturing knowledge. The main task performed in these systems is using inductive methods to the given values of attributes of an unknown object to determine appropriate classification according to decision tree rules. In the past, ID3 was the most used algorithm in this area . This algorithm is introduced by Quinlan, using information theory to determine the most informative attribute. A disadvantage of decision tree is its instability. Decision tree is recognized as highly unstable classifier. The structure of the decision tree may be entirely different if some things change in the dataset. To overcome this problem, some scholars have suggested Fuzzy Decision Tree (e.g. FuzzyID3) by utilizing the fuzzy set theory to describe the connected degree of attribute values, which can precisely distinguish the deference of subordinate relations between different examples and every attribute values. After some years PFID3 was also introduced which was called as probabilistic fuzzy ID3. In this paper, a comparative study on ID3, FID3 and PFID3 is done.