Decision Tree Modeling in Classification

Predicting ore content throughout a machine learning procedure – An Sn-W enrichment case study

“…Given a set of continuous or categorical independent variables, classification and regression trees (CART) are non-parametric techniques that can explain the response of a dependent variable (Breiman et al, 1984;Moisen, 2008). The initial data are split into subgroups, then the variance of these subgroups is minimized until they reach homogeneity (Brown and Myles, 2009).…”

Prediction of mechanical strength of cork under compression using machine learning techniques

“…The model divides the initial database into subgroups with the aim of finding homogeneous groups using variance minimization algorithms [37]. Depending on the type of tree, the common optimization algorithms are: mean-squared error (MSE) for regression tress; Gini's diversity index, twoing rule or deviance for classification trees.…”

Possible Decision Rules to Allocate Quotas and Reservations to Ensure Equity for Nepalese Poor

“…Pruning and cross-validation simplifies the growth of tree and improve the stability and predictive accuracy (Kim & Upneja, 2014). An n -fold (with 10 3 − = n ) cross validation is a common empirical approach to optimize tree growth (Brown & Myles, 2009). It is a re-sampling technique that uses multiple random training and validating sub samples.…”

“…CART algorithm uses Gini index 2 measure as the splitting criteria and does only binary splitting (Bozkir & Sezer, 2011). Gini index generalizes the variance impurity -the variance of a distribution associated with two classes (Brown & Myles, 2009). This is a recursive process as it repeats until a predefined measure of homogeneity/purity (or other measure of completion) is satisfied.…”