Decision Tree Learning Algorithm

AKA: Supervised Decision Tree Learning Algorithm, Tree Induction Algorithm.
Context:
- It can be implemented in a Decision Tree Training System (that can solve a Decision Tree Training Task).
- It is a Symbolic Learning Algorithm.
- It is a Discriminative Learning Algorithm.
- It can make use of a Decision Tree Splitting Criterion.
- It can range from being a Fully-Supervised Decision Tree Learning Algorithm to being a Semi-Supervised Decision Tree Learning Algorithm.
- It can be an Incremental Decision Tree Algorithm
Example(s):
Counter-Example(s):
See: Instance-based Learning Algorithm, Tree Learning Algorithm.

References

(Wikipedia, 2011) ⇒ http://en.wikipedia.org/wiki/Decision_tree_learning
- Decision tree learning, used in statistics, data mining and machine learning, uses a decision tree as a predictive model which maps observations about an item to conclusions about the item's target value. More descriptive names for such tree models are classification trees or regression trees. In these tree structures, leaves represent classifications and branches represent conjunctions of features that lead to those classifications. In decision analysis, a decision tree can be used to visually and explicitly represent decisions and decision making. In data mining, a decision tree describes data but not decisions; rather the resulting classification tree can be an input for decision making. This page deals with decision trees in data mining.

(Wilson, 2008a) ⇒ Bill Wilson. (2008). "The Machine Learning Dictionary for COMP9414." University of New South Wales, Australia.
- tree induction algorithm: This article describes the basic tree induction algorithm used by ID3 and successors. The basic idea is to pick an attribute A with values \(a_1, a_2, ..., a_r\), split the training instances into subsets \(S_{a1}, S_{a2}, ..., S_{ar}\) consisting of those instances that have the corresponding attribute value. Then if a subset has only instances in a single class, that part of the tree stops with a leaf node labelled with the single class. If not, then the subset is split again, recursively, using a different attribute. This leaves the question of how to choose the best attribute to split on at any branch node. This issue is handled in the article on splitting criterion in ID3.

(Quinlan, 1996) ⇒ J. Ross Quinlan. (1996). "Improved Use of Continuous Attributes in C4.5." In: Journal of Artificial Intelligence Research, 4.