Year: 1994
Title: Exploring the Decision Forest: an empirical investigation of Occam's razor in decision tree induction
Journal: Journal of Artificial Intelligence Research, 1, 257-275
Comments: Using masively parallel computing techniques they collect all the decision trees that correctly classify a set of training data and then asses these on unseen data from the same source. The shortest trees are not the best, the slightly deeper ones do better, thus contradicting Occam's Razor.
Keywords: SIZE, SIMPLICITY, DEPTH,
| Top | Authors | Journals | Years | Keywords | Search | New | Comments |