An Analysis of Ensemble Pruning Techniques Based on Ordered Aggregation

Several pruning strategies that can be used to reduce the size and increase the accuracy of bagging ensembles are analyzed. These heuristics select subsets of complementary classifiers that, when combined, can perform better than the whole ensemble. The pruning methods investigated are based on modifying the order of aggregation of classifiers in the ensemble. In the original bagging algorithm, the order of aggregation is left unspecified. When this order is random, the generalization error typically decreases as the number of classifiers in the ensemble increases. If an appropriate ordering for the aggregation process is devised, the generalization error reaches a minimum at intermediate numbers of classifiers. This minimum lies below the asymptotic error of bagging. Pruned ensembles are obtained by retaining a fraction of the classifiers in the ordered ensemble. The performance of these pruned ensembles is evaluated in several benchmark classification tasks under different training conditions. The results of this empirical investigation show that ordered aggregation can be used for the efficient generation of pruned ensembles that are competitive, in terms of performance and robustness of classification, with computationally more costly methods that directly select optimal or near-optimal subensembles.     

Attribute bagging: improving accuracy of classifier ensembles by using random feature subsets

We present attribute bagging (AB), a technique for improving the accuracy and stability of classifier ensembles induced using random subsets of features. AB is a wrapper method that can be used with any learning algorithm. It establishes an appropriate attribute subset size and then randomly selects subsets of features, creating projections of the training set on which the ensemble classifiers are built. The induced classifiers are then used for voting. This article compares the performance of our AB method with bagging and other algorithms on a hand-pose recognition dataset. It is shown that AB gives consistently better results than bagging, both in accuracy and stability. The performance of ensemble voting in bagging and the AB method as a function of the attribute subset size and the number of voters for both weighted and unweighted voting is tested and discussed. We also demonstrate that ranking the attribute subsets by their classification accuracy and voting using only the best subsets further improves the resulting performance of the ensemble.     

Preprocessed dynamic classifier ensemble selection for highly imbalanced drifted data streams

This work aims to connect two rarely combined research directions, i.e., non-stationary data stream classification and data analysis with skewed class distributions. We propose a novel framework employing stratified bagging for training base classifiers to integrate data preprocessing and dynamic ensemble selection methods for imbalanced data stream classification. The proposed approach has been evaluated based on computer experiments carried out on 135 artificially generated data streams with various imbalance ratios, label noise levels, and types of concept drift as well as on two selected real streams. Four preprocessing techniques and two dynamic selection methods, used on both bagging classifiers and base estimators levels, were considered. Experimentation results showed that, for highly imbalanced data streams, dynamic ensemble selection coupled with data preprocessing could outperform online and chunk-based state-of-art methods.     

Improving bagging performance through multi-algorithm ensembles

Working as an ensemble method that establishes a committee of classifiers first and then aggregates their outcomes through majority voting, bagging has attracted considerable research interest and been applied in various application domains. It has demonstrated several advantages, but in its present form, bagging has been found to be less accurate than some other ensemble methods. To unlock its power and expand its user base, we propose an approach that improves bagging through the use of multi-algorithm ensembles. In a multi-algorithm ensemble, multiple classification algorithms are employed. Starting from a study of the nature of diversity, we show that compared to using different training sets alone, using heterogeneous algorithms together with different training sets increases diversity in ensembles, and hence we provide a fundamental explanation for research utilizing heterogeneous algorithms. In addition, we partially address the problem of the relationship between diversity and accuracy by providing a non-linear function that describes the relationship between diversity and correlation. Furthermore, after realizing that the bootstrap procedure is the exclusive source of diversity in bagging, we use heterogeneity as another source of diversity and propose an approach utilizing heterogeneous algorithms in bagging. For evaluation, we consider several benchmark data sets from various application domains. The results indicate that, in terms of F1-measure, our approach outperforms most of the other state-of-the-art ensemble methods considered in experiments and, in terms of mean margin, our approach is superior to all the others considered in experiments.     

Overfitting cautious selection of classifier ensembles with genetic algorithms

Information fusion research has recently focused on the characteristics of the decision profiles of ensemble members in order to optimize performance. These characteristics are particularly important in the selection of ensemble members. However, even though the control of overfitting is a challenge in machine learning problems, much less work has been devoted to the control of overfitting in selection tasks. The objectives of this paper are: (1) to show that overfitting can be detected at the selection stage; and (2) to present strategies to control overfitting. Decision trees and k nearest neighbors classifiers are used to create homogeneous ensembles, while single- and multi-objective genetic algorithms are employed as search algorithms at the selection stage. In this study, we use bagging and random subspace methods for ensemble generation. The classification error rate and a set of diversity measures are applied as search criteria. We show experimentally that the selection of classifier ensembles conducted by genetic algorithms is prone to overfitting, especially in the multi-objective case. In this study, the partial validation, backwarding and global validation strategies are tailored for classifier ensemble selection problem and compared. This comparison allows us to show that a global validation strategy should be applied to control overfitting in pattern recognition systems involving an ensemble member selection task. Furthermore, this study has helped us to establish that the global validation strategy can be used to measure the relationship between diversity and classification performance when diversity measures are employed as single-objective functions.     

Ensembling local learners through multimodal perturbation.

Ensemble learning algorithms train multiple component learners and then combine their predictions. In order to generate a strong ensemble, the component learners should be with high accuracy as well as high diversity. A popularly used scheme in generating accurate but diverse component learners is to perturb the training data with resampling methods, such as the bootstrap sampling used in bagging. However, such a scheme is not very effective on local learners such as nearest-neighbor classifiers because a slight change in training data can hardly result in local learners with big differences. In this paper, a new ensemble algorithm named Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR) is proposed for building ensembles of local learners, which utilizes multimodal perturbation to help generate accurate but diverse component learners. In detail, FASBIR employs the perturbation on the training data with bootstrap sampling, the perturbation on the input attributes with attribute filtering and attribute subspace selection, and the perturbation on the learning parameters with randomly configured distance metrics. A large empirical study shows that FASBIR is effective in building ensembles of nearest-neighbor classifiers, whose performance is better than that of many other ensemble algorithms.     

Empirical analysis of support vector machine ensemble classifiers

Ensemble classification – combining the results of a set of base learners – has received much attention in the machine learning community and has demonstrated promising capabilities in improving classification accuracy. Compared with neural network or decision tree ensembles, there is no comprehensive empirical research in support vector machine (SVM) ensembles. To fill this void, this paper analyses and compares SVM ensembles with four different ensemble constructing techniques, namely bagging, AdaBoost, Arc-X4 and a modified AdaBoost. Twenty real-world data sets from the UCI repository are used as benchmarks to evaluate and compare the performance of these SVM ensemble classifiers by their classification accuracy. Different kernel functions and different numbers of base SVM learners are tested in the ensembles. The experimental results show that although SVM ensembles are not always better than a single SVM, the SVM bagged ensemble performs as well or better than other methods with a relatively higher generality, particularly SVMs with a polynomial kernel function. Finally, an industrial case study of gear defect detection is conducted to validate the empirical analysis results.     

An assessment of the effectiveness of a rotation forest ensemble for land-use and land-cover mapping

Increasing the accuracy of thematic maps produced through the process of image classification has been a hot topic in remote sensing. For this aim, various strategies, classifiers, improvements, and their combinations have been suggested in the literature. Ensembles that combine the prediction of individual classifiers with weights based on the estimated prediction accuracies are strategies aiming to improve the classifier performances. One of the recently introduced ensembles is the rotation forest, which is based on the idea of building accurate and diverse classifiers by applying feature extraction to the training sets and then reconstructing new training sets for each classifier. In this study, the effectiveness of the rotation forest was investigated for decision trees in land-use and land-cover (LULC) mapping, and its performance was compared with performances of the six most widely used ensemble methods. The results were verified for the effectiveness of the rotation forest ensemble as it produced the highest classification accuracies for the selected satellite data. When the statistical significance of differences in performances was analysed using McNemar's tests based on normal and chi-squared distributions, it was found that the rotation forest method outperformed the bagging, Diverse Ensemble Creation by Oppositional Relabelling of Artificial Training Examples (DECORATE), and random subspace methods, whereas the performance differences with the other ensembles were statistically insignificant.     

Incremental construction of classifier and discriminant ensembles

We discuss approaches to incrementally construct an ensemble. The first constructs an ensemble of classifiers choosing a subset from a larger set, and the second constructs an ensemble of discriminants, where a classifier is used for some classes only. We investigate criteria including accuracy, significant improvement, diversity, correlation, and the role of search direction. For discriminant ensembles, we test subset selection and trees. Fusion is by voting or by a linear model. Using 14 classifiers on 38 data sets, incremental search finds small, accurate ensembles in polynomial time. The discriminant ensemble uses a subset of discriminants and is simpler, interpretable, and accurate. We see that an incremental ensemble has higher accuracy than bagging and random subspace method; and it has a comparable accuracy to AdaBoost, but fewer classifiers.     

A theoretical analysis of bagging as a linear combination of classifiers

We apply an analytical framework for the analysis of linearly combined classifiers to ensembles generated by bagging. This provides an analytical model of bagging misclassification probability as a function of the ensemble size, which is a novel result in the literature. Experimental results on real data sets confirm the theoretical predictions. This allows us to derive a novel and theoretically grounded guideline for choosing bagging ensemble size. Furthermore, our results are consistent with explanations of bagging in terms of classifier instability and variance reduction, support the optimality of the simple average over the weighted average combining rule for ensembles generated by bagging, and apply to other randomization-based methods for constructing classifier ensembles. Although our results do not allow to compare bagging misclassification probability with the one of an individual classifier trained on the original training set, we discuss how the considered theoretical framework could be exploited to this aim.     

基于精简随机子空间的多生成树集成一类分类算法

由于高维数据通常存在冗余和噪声,在其上直接构造覆盖模型不能充分反映数据的分布信息,导致分类器性能下降.为此提出一种基于精简随机子空间多树集成分类方法.该方法首先生成多个随机子空间,并在每个子空间上构造独立的最小生成树覆盖模型.其次对每个子空间上构造的分类模型进行精简处理,通过一个评估准则(AUC值),对生成的一类分类器进行精简.最后均值合并融合这些分类器为一个集成分类器.实验结果表明,与其它直接覆盖分类模型和bagging算法相比,多树集成覆盖分类器具有更高的分类正确率.     

An Evolutionary Algorithm Approach to Optimal Ensemble Classifiers for DNA Microarray Data Analysis

In general, the analysis of microarray data requires two steps: feature selection and classification. From a variety of feature selection methods and classifiers, it is difficult to find optimal ensembles composed of any feature-classifier pairs. This paper proposes a novel method based on the evolutionary algorithm (EA) to form sophisticated ensembles of features and classifiers that can be used to obtain high classification performance. In spite of the exponential number of possible ensembles of individual feature-classifier pairs, an EA can produce the best ensemble in a reasonable amount of time. The chromosome is encoded with real values to decide the weight for each feature-classifier pair in an ensemble. Experimental results with two well-known microarray datasets in terms of time and classification rate indicate that the proposed method produces ensembles that are superior to individual classifiers, as well as other ensembles optimized by random and greedy strategies.     

Statistical Instance-Based Pruning in Ensembles of Independent Classifiers

The global prediction of a homogeneous ensemble of classifiers generated in independent applications of a randomized learning algorithm on a fixed training set is analyzed within a Bayesian framework. Assuming that majority voting is used, it is possible to estimate with a given confidence level the prediction of the complete ensemble by querying only a subset of classifiers. For a particular instance that needs to be classified, the polling of ensemble classifiers can be halted when the probability that the predicted class will not change when taking into account the remaining votes is above the specified confidence level. Experiments on a collection of benchmark classification problems using representative parallel ensembles, such as bagging and random forests, confirm the validity of the analysis and demonstrate the effectiveness of the instance-based ensemble pruning method proposed.     

Applying tree-based ensemble algorithms to the classification of ecological zones using multi-temporal multi-source remote-sensing data

The decision tree method has grown fast in the past two decades and its performance in classification is promising. The tree-based ensemble algorithms have been used to improve the performance of an individual tree. In this study, we compared four basic ensemble methods, that is, bagging tree, random forest, AdaBoost tree and AdaBoost random tree in terms of the tree size, ensemble size, band selection (BS), random feature selection, classification accuracy and efficiency in ecological zone classification in Clark County, Nevada, through multi-temporal multi-source remote-sensing data. Furthermore, two BS schemes based on feature importance of the bagging tree and AdaBoost tree were also considered and compared. We conclude that random forest or AdaBoost random tree can achieve accuracies at least as high as bagging tree or AdaBoost tree with higher efficiency; and although bagging tree and random forest can be more efficient, AdaBoost tree and AdaBoost random tree can provide a significantly higher accuracy. All ensemble methods provided significantly higher accuracies than the single decision tree. Finally, our results showed that the classification accuracy could increase dramatically by combining multi-temporal and multi-source data set.     

A comparison of decision tree ensemble creation techniques

We experimentally evaluate bagging and seven other randomization-based approaches to creating an ensemble of decision tree classifiers. Statistical tests were performed on experimental results from 57 publicly available data sets. When cross-validation comparisons were tested for statistical significance, the best method was statistically more accurate than bagging on only eight of the 57 data sets. Alternatively, examining the average ranks of the algorithms across the group of data sets, we find that boosting, random forests, and randomized trees are statistically significantly better than bagging. Because our results suggest that using an appropriate ensemble size is important, we introduce an algorithm that decides when a sufficient number of classifiers has been created for an ensemble. Our algorithm uses the out-of-bag error estimate, and is shown to result in an accurate ensemble for those methods that incorporate bagging into the construction of the ensemble     

Combining bagging,boosting, rotation forest and random subspace methods

Bagging, boosting, rotation forest and random subspace methods are well known re-sampling ensemble methods that generate and combine a diversity of learners using the same learning algorithm for the base-classifiers. Boosting and rotation forest algorithms are considered stronger than bagging and random subspace methods on noise-free data. However, there are strong empirical indications that bagging and random subspace methods are much more robust than boosting and rotation forest in noisy settings. For this reason, in this work we built an ensemble of bagging, boosting, rotation forest and random subspace methods ensembles with 6 sub-classifiers in each one and then a voting methodology is used for the final prediction. We performed a comparison with simple bagging, boosting, rotation forest and random subspace methods ensembles with 25 sub-classifiers, as well as other well known combining methods, on standard benchmark datasets and the proposed technique had better accuracy in most cases.     

Novel ensemble methods for regression via classification problems

Regression via classification (RvC) is a method in which a regression problem is converted into a classification problem. A discretization process is used to covert continuous target value to classes. The discretized data can be used with classifiers as a classification problem. In this paper, we use a discretization method, Extreme Randomized Discretization (ERD), in which bin boundaries are created randomly to create ensembles. We present two ensemble methods for RvC problems. We show theoretically that the proposed ensembles for RvC perform better than RvC with the equal-width discretization method. We also show the superiority of the proposed ensemble methods experimentally. Experimental results suggest that the proposed ensembles perform competitively to the method developed specifically for regression problems.     

A comparative study of classifier ensembles for bankruptcy prediction

The aim of bankruptcy prediction in the areas of data mining and machine learning is to develop an effective model which can provide the higher prediction accuracy. In the prior literature, various classification techniques have been developed and studied, in/with which classifier ensembles by combining multiple classifiers approach have shown their outperformance over many single classifiers. However, in terms of constructing classifier ensembles, there are three critical issues which can affect their performance. The first one is the classification technique actually used/adopted, and the other two are the combination method to combine multiple classifiers and the number of classifiers to be combined, respectively. Since there are limited, relevant studies examining these aforementioned disuses, this paper conducts a comprehensive study of comparing classifier ensembles by three widely used classification techniques including multilayer perceptron (MLP) neural networks, support vector machines (SVM), and decision trees (DT) based on two well-known combination methods including bagging and boosting and different numbers of combined classifiers. Our experimental results by three public datasets show that DT ensembles composed of 80–100 classifiers using the boosting method perform best. The Wilcoxon signed ranked test also demonstrates that DT ensembles by boosting perform significantly different from the other classifier ensembles. Moreover, a further study over a real-world case by a Taiwan bankruptcy dataset was conducted, which also demonstrates the superiority of DT ensembles by boosting over the others.     

An exploratory study of mono and multi-objective metaheuristics to ensemble of classifiers

This paper performs an exploratory study of the use of metaheuristic optimization techniques to select important parameters (features and members) in the design of ensemble of classifiers. In order to do this, an empirical investigation, using 10 different optimization techniques applied to 23 classification problems, will be performed. Furthermore, we will analyze the performance of both mono and multi-objective versions of these techniques, using all different combinations of three objectives, classification error as well as two important diversity measures to ensembles, which are good and bad diversity measures. Additionally, the optimization techniques will also have to select members for heterogeneous ensembles, using k-NN, Decision Tree and Naive Bayes as individual classifiers and they are all combined using the majority vote technique. The main aim of this study is to define which optimization techniques obtained the best results in the context of mono and multi-objective as well as to provide a comparison with classical ensemble techniques, such as bagging, boosting and random forest. Our findings indicated that three optimization techniques, Memetic, SA and PSO, provided better performance than the other optimization techniques as well as traditional ensemble generator (bagging, boosting and random forest).     

Two-level classifier ensembles for credit risk assessment

Many techniques have been proposed for credit risk assessment, from statistical models to artificial intelligence methods. During the last few years, different approaches to classifier ensembles have successfully been applied to credit scoring problems, demonstrating to be generally more accurate than single prediction models. The present paper goes one step beyond by introducing composite ensembles that jointly use different strategies for diversity induction. Accordingly, the combination of data resampling algorithms (bagging and AdaBoost) and attribute subset selection methods (random subspace and rotation forest) for the construction of composite ensembles is explored with the aim of improving the prediction performance. The experimental results and statistical tests show that this new two-level classifier ensemble constitutes an appropriate solution for credit scoring problems, performing better than the traditional single ensembles and very significantly better than individual classifiers.