Knowledge Acquisition from Examples Via Multiple Models

By Pedro M. Domingos. 1997

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

Interpreting tree ensembles with inTrees (2014, methodology): “To gain interpretability for tree ensembles, methods were proposed to approximate tree ensembles by single decision trees [7,12,20].”
Model compression (2006, background): “CMM is a meta-learner that learns one single model from a bagged ensemble of the same type of models [8].”
Tree Induction for Probability-Based Ranking (2003, methodology): “4 tree may bepreferable, or a method like CMM (Domingos, 1997) (whi h produ esa single-tree approximation of the ensemble) may be useful.8.”
A Survey of Methods for Scaling Up Inductive Algorithms (1999, background): “A quite different approach to creating comprehensible classi-ﬁers from ensembles is taken by Craven (1996), by Domingos (1997), and by Guo and Sutiwaraphun (1998).”
CTC : an alternative to extract explanation from bagging (2007, methodology): “, where it is not enough to obtain the right classification and the users wish to gain insight into the domain [5].”
An Analysis of Ensemble Pruning Techniques Based on Ordered Aggregation (2009, methodology): “Combined Multiple Models [34] builds a single classifier using a training set that, besides the original training examples, includes synthetic examples labeled by the ensemble.”
Creating diversity in ensembles using artificial data (2005, unknown): “Another application of artificial examples for ensembles is Combined Multiple Models (CMMs) (Domingos, 1997).”
A survey of evolutionary algorithms for data mining and knowledge discovery (2003, background): “Actually, generating an ensemble of classifiers is a relatively recent trend in machine learning when our primary goal is to maximize predictive accuracy, since it has been shown that in several cases an ensemble of classifiers has a better predictive accuracy than a single classifier [58], [15].”