Trading Accuracy for Simplicity in Decision Trees

By Marko Bohanec, Ivan Bratko. 1994

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Survey of Methods for Explaining Black Box Models (2018, methodology): “In the decision tree literature, the problems of over-fitting the training data and of “trading accuracy for simplicity” [10] have been addressed by a class of post-processing algorithms called”
Evolutionary stratified training set selection for extracting classification rules with trade off precision-interpretability (2007, background): “This situation is increased when the size of the learning set is large, so decision trees size is increased considerably [17] [18] [19].”
Data mining with an ant colony optimization algorithm (2002, background): “tually, there are several classification-rule discovery algorithms that were explicitly designed to improve rule set simplicity, even at the expense of reducing the predictive accuracy [1], [3], [4].”
Subtree Replacement in Decision Tree Simplification (2012, methodology): “We study the simplification operator of subtree replacement, also known as grafting, originally implemented in the C4.5 system.”
Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey (1998, background): “Use of dynamic programming to prune trees optimally and e ciently has been explored in [25].”
A statistical approach to rule learning (2006, background): “, 2004) and decision tree learning (for example Bohanec and Bratko, 1994). It is also known that simple classifiers can feature high predictive accuracy. For instance, Holte (1993) showed that for some real-world data sets, decision stump classifiers can have predictive accuracies that are as high a”
Top-down induction of decision trees classifiers - a survey (2005, background): “Another key motivation of pruning is ”trading accuracy for simplicity” as presented by Bratko and Bohanec [39].”
Evolutionary Algorithms for Solving Multi-Objective Problems (2002, unknown): “Results are compared against feature subset selection (Almuallin and Dietterich, 1991) (FSS), ID3, and OPT (Bohanec and Bratko, 1994).”
Interacting meaningfully with machine learning systems: Three experiments (2009, methodology): “Examples of these approaches include pruning to simplify decision trees (Bohanec and Bratko 1994; Cunningham et al. 1996) and inducing comprehensible models, such as decision trees (Craven 1996), from neural networks, which are notoriously difficult for a human to understand.”
An experimental evaluation of simplicity in rule learning (2008, background): “In machine learning, simplicity in decision tree learning has been investigated by Bohanec and Bratko (24).”
Model complexity of deep learning: a survey (2021, background): “A series of studies [16, 61] investigate tree optimization based on the accuracy-complexity tradeoff.”