Appropriate Problems for Decision Tree Learning :
However remember there that is a skilled job in "AI" to choose exactly the right learning representation ormethod for a particular learning task. Moreover as elaborated by Tom Mitchell and decision tree learning is the best suited to problems with some of these characteristics as:
• Here the background concepts describe the examples in terms of attribute-value pairs so the values for each attribute range over finitely many fixed possibilities, and
• There the concept to be learned like Mitchell calls it the target function has discrete values.
• Through disjunctive descriptions might be required in the answer.
However just in addition to this there decision tree learning is robust to errors in the data. So in particular there it will function well in the light of (i) the errors in the classification instances provided (ii) the errors in the attribute-value pairs provided (iii) and missing values for certain attributes for certain examples.