Weighted Scoring in Geometric Space for Decision Tree Ensemble

Abstract

In order to improve the classification performance of a single classification model, Multiple Classifier Systems (MCS) are used. One of the most common techniques utilizing multiple decision trees is the random forest, where diversity between base classifiers is obtained by bagging the training dataset. In this paper, we propose the algorithm that uses horizontal partitioning the learning set and uses decision trees as base models to obtain decision regions. In the proposed approach feature space is divided into disjoint subspace. Additionally, the location of the subspace centroids, as well as the size and location of decision regions, are used in order to determine the weights needed in the last process of creating MCS, i.e. in the integration phase. The proposed algorithm was evaluated employing multiple open-source benchmarking datasets, compared using accuracy and Matthews correlation coefficient performance measures with two existing MCS methods-random forest and majority voting. The statistical analysis confirms an improvement in recognition compared to the random forest. In addition, we proved that for infinitely dense space division proposed algorithm is equivalent to majority voting. INDEX TERMS Decision tree, ensemble classifier, majority voting, multiple classifier system, random forest.

Decision Tree Integration Using Dynamic Regions of Competence

“…The proposed method is based on previous works of authors, but suggests a slightly different approach [ 23 , 24 ]. While the cited articles used static division into regions of competence, this paper presents an algorithm with a dynamic approach.…”

“…The authors have studied and proved the effectiveness of an integration algorithm based on averaging and taking median of values of the decision boundary in the SVM classifiers [ 22 ]. Next, two algorithms for decision trees were proposed and evaluated [ 23 , 24 ]. They have proven themselves to provide better classification quality and ease of use than referential methods.…”

Classification of Contaminated Insulators Using k-Nearest Neighbors Based on Computer Vision

“…After evaluating the best configuration of the model, a benchmarking with the decision tree [68], ensemble [69][70][71], support vector machine (SVM) [72][73][74], and the multilayer perceptron models are presented. The pictures of the insulators were taken before the measurement of the NSDD, so that there was no influence from the operator on the contamination; if the contamination did not meet the requirement of IEC 60815 (Annex C) [52], the process was repeated from the beginning.…”

Classification of Contaminated Insulators Using k-Nearest Neighbors Based on Computer Vision

“…After evaluating the best configuration of the model, a benchmarking with the decision tree [48], ensemble [49,50] and support vector machine (SVM) [51][52][53] models is presented. The pictures of the insulators were taken before the measurement of the NSDD, so that there was no influence from the operator on the contamination, if the contamination did not meet the requirement of IEC 60815 (Annex C) [32] the process would be repeated since the start.…”