Learning With Continuous Classes

By J. Ross Quinlan. 1992

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

Toward Interpretable Deep Reinforcement Learning with Linear Model U-Trees (2018, background): “M5 [15] is a tree training algorithm with more generalization ability.”
Is Combining Classifiers with Stacking Better than Selecting the Best One? (2004, methodology): “M5′ (Wang & Witten, 1997), a re-implementation of M5 (Quinlan, 1992) included in the data mining suite Weka (Witten & Frank, 1999), is used to induce the model trees at the meta level.”
Comparing various machine learning approaches in modeling the dynamic viscosity of CuO/water nanofluid (2019, methodology): “In the first method, the pruning process prevents the generation of extra branches while in the second approach, pruning is done after the formation of the tree [94].”
Learning model trees from evolving data streams (2010, methodology): “1998) from R (default settings), and M5′ (Quinlan 1992) from WEKA (default settings) with the regression-tree option (M5′RT).”
Recent advances in decision trees: an updated survey (2022, methodology): “Often, researchers tackle this problem either by growing a standard tree and adding models in post-processing (Quinlan 1992), or by inducing a model for each internal node and then using this model to warm-start the node’s children, therefore reducing the cost of training a new model (Landwehr et…”
Humanoid robots learning to walk faster: from the real world to simulation and back (2013, background): “A detailed description of the M5 algorithm can be found in [22].”
Do we need hundreds of classifiers to solve real world classification problems? (2014, methodology): “, 1998) binarizes each class and learns its corresponding M5P tree/rule regression model (Quinlan, 1992), with at least 4 training patterns per leaf.”
Evaporation modelling by heuristic regression approaches using only temperature data (2019, methodology): “The DT works with respect to two conditions: maximize the information and minimize the error in the branches of the tree (Quinlan 1992, Wang and Witten 1997).”
M5 model tree application in daily river flow forecasting in Sohu Stream, Turkey (2013, methodology): “The M5 algorithm is used for inducing a model tree [24] in the present study.”
pre: An R Package for Fitting Prediction Rule Ensembles (2017, methodology): “Weka’s (Hall, Frank, Holmes, Pfahringer, Reutemann, and Witten 2009) sub-package classi-ﬁers.rules implements several algorithms for deriving PREs: JRip (implementing the RIPPER algorithm of Cohen 1995), M5Rules (Quinlan 1992; Holmes, Hall, and Frank 1999) and PART (Frank and Witten 1998).”
On digital soil mapping (2003, background): “5 (Quinlan, 1992) and its later version See5 (Rulequest Research, 2000).”