Simplifying Decision Trees

By J. R. Quinlan. 1987

Many systems have been developed for constructing decision trees from collections of examples. Although the decision trees generated by these methods are accurate and efficient, they often suffer the disadvantage of excessive complexity and are therefore incomprehensible to experts. It is questionable whether opaque structures of this kind can be described as knowledge, no matter how well they function. This paper discusses techniques for simplifying decision trees while retaining their accuracy. Four methods are described, illustrated, and compared on a test-bed of decision trees from a variety of domains.

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

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, neutral): “Decision trees are hierarchical structures for decision making used to support regression and classi-ﬁcation problems [132,180] .”
QUOTIENT: Two-Party Secure Neural Network Training and Prediction (2019, neutral): “Five of these are privacy sensitive in nature, as they include health datasets—Thyroid [45], Breast cancer [8], MotionSense [38], Skin cancer MNIST [53] and a financial credit dataset—German credit [12].”
Techniques for interpretable machine learning (2018, background): “tures to have monotonic relations with the prediction. Similarly, decision trees are pruned by replacing subtrees with leaves to encourage long and deep trees rather than wide and more balanced trees [29]. These constraints make a model simpler and could increase the model’s comprehensibility by use”
A review of machine learning in processing remote sensing data for mineral exploration (2021, methodology): “The algorithm is non-parametric and can efficiently deal with large, complicated datasets without imposing a complicated parametric structure (Song and Ying, 2015; Quinlan, 1987).”
A novel evolutionary data mining algorithm with applications to churn prediction (2003, background): “3 The Credit Card Database The credit card database [35] contains data about credit card applications.”
TMVA - Toolkit for Multivariate Data Analysis (2007, methodology): “ge. This is because apparently insigniﬁcant splits can nevertheless lead to go od splits further down the tree. TMVA currently implements two tree pruning algorithms. • For the expected error pruning [19] all leaf nodes for which the statistical error estimates of the parent nodes are smaller than t”
Collateral, relationship lending and family firms (2010, background): “This process is continued until further subdivision does not cause a significant improvement of the model ( Quinlan 1993 ).”
An Optimal Constrained Pruning Strategy for Decision Trees (2009, methodology): “Besides, the focus in Zhang and Huei-chuen (2005) is to use their IP formulation to establish a new optimality proof of the bottom-up pruning method due to Quinlan (1988), as opposed to actually modeling and solving the types of constrained pruning problems addressed herein.”
Learning logical definitions from relations (1990, background): “In such circumstances, many studies have found that simple, approximate concepts can be more valuable than overspecified rules that "fit the noise." In the case of classification knowledge expressed as decision trees or as rules, improved performance on unseen cases can be obtained by "pruning" (Bre”
A survey of the applications of text mining in financial domain (2016, methodology): “3, which consists of the regular classification methods i.e. DT (Quinlan, 1987), SVM (Cortes and Vapnik, 1995), NB (Domingos and Pazzani 1997), and k-NN (Cover and Hart 1967).”