The random subspace method for constructing decision forests

Much of previous attention on decision trees focuses on the splitting criteria and optimization of tree sizes. The dilemma between overfitting and achieving maximum accuracy is seldom resolved. A method to construct a decision tree based classifier is proposed that maintains highest accuracy on training data and improves on generalization accuracy as it grows in complexity. The classifier consists of multiple trees constructed systematically by pseudorandomly selecting subsets of components of the feature vector, that is, trees constructed in randomly chosen subspaces. The subspace method is compared to single-tree classifiers and other forest construction methods by experiments on publicly available datasets, where the method's superiority is demonstrated. We also discuss independence between trees in a forest and relate that to the combined classification accuracy     

A noise-detection based AdaBoost algorithm for mislabeled data

Noise sensitivity is known as a key related issue of AdaBoost algorithm. Previous works exhibit that AdaBoost is prone to be overfitting in dealing with the noisy data sets due to its consistent high weights assignment on hard-to-learn instances (mislabeled instances or outliers). In this paper, a new boosting approach, named noise-detection based AdaBoost (ND-AdaBoost), is exploited to combine classifiers by emphasizing on training misclassified noisy instances and correctly classified non-noisy instances. Specifically, the algorithm is designed by integrating a noise-detection based loss function into AdaBoost to adjust the weight distribution at each iteration. A k-nearest-neighbor (k-NN) and an expectation maximization (EM) based evaluation criteria are both constructed to detect noisy instances. Further, a regeneration condition is presented and analyzed to control the ensemble training error bound of the proposed algorithm which provides theoretical support. Finally, we conduct some experiments on selected binary UCI benchmark data sets and demonstrate that the proposed algorithm is more robust than standard and other types of AdaBoost for noisy data sets.     

Building bagging on critical instances

The ensemble method is a powerful data mining paradigm, which builds a classification model by integrating multiple diversified component learners. Bagging is one of the most successful ensemble methods. It is made of bootstrap-inspired classifiers and uses these classifiers to get an aggregated classifier. However, in bagging, bootstrapped training sets become more and more similar as redundancy is increasing. Besides redundancy, any training set is usually subject to noise. Moreover, the training set might be imbalanced. Thus, each training instance has a different impact on the learning process. This paper explores some properties of the ensemble margin and its use in improving the performance of bagging. We introduce a new approach to measure the importance of training data in learning, based on the margin theory. Then, a new bagging method concentrating on critical instances is proposed. This method is more accurate than bagging and more robust than boosting. Compared to bagging, it reduces the bias while generally keeping the same variance. Our findings suggest that (a) examples with low margins tend to be more critical for the classifier performance; (b) examples with higher margins tend to be more redundant; (c) misclassified examples with high margins tend to be noisy examples. Our experimental results on 15 various data sets show that the generalization error of bagging can be reduced up to 2.5% and its resilience to noise strengthened by iteratively removing both typical and noisy training instances, reducing the training set size by up to 75%.     

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.     

适于数据流组合分类的直推学习方法

在进行组合决策时,已有的组合分类方法需要对多个组合分类器均有效的公共已知标签训练样本。为了解决在没有已知标签样本的情况下数据流组合分类决策问题,提出一种基于约束学习的数据流组合分类器的融合策略。在判定测试样本上的决策时,根据直推学习理论设计满足每一个局部分类器约束度量的方法,保证了约束的可行性,解决了分布式分类聚集时最大熵的直推扩展问题。测试数据集上的实验证明,与已有的直推学习方法相比,此方法可以获得更好的决策精度,可以应用于数据流组合分类的融合。     

一种辅以强规则学习的双层文本分类模型

随着基于机器学习的文本自动分类方法成为主流分类技术，基于机器学习的文本分类方法往往忽视了对规则分类方法的有效运用。该文将基于规则的分类思想和基于机器学习的分类方法有机地结合起来，把规则判别看作一个分量分类器，提出了一种辅以规则补充的双层文本分类模型和一种优化的分类规则学习算法。根据该方法设计并实现了一个基于规则和N-Gram统计分类相结合的双层分类器，进行了双层分类模型与单独的N-Gram分类模型的实验，结果表明辅以规则补充的双层分类器具有更好的分类性能。     

Boosting Algorithms for Parallel and Distributed Learning

The growing amount of available information and its distributed and heterogeneous nature has a major impact on the field of data mining. In this paper, we propose a framework for parallel and distributed boosting algorithms intended for efficient integrating specialized classifiers learned over very large, distributed and possibly heterogeneous databases that cannot fit into main computer memory. Boosting is a popular technique for constructing highly accurate classifier ensembles, where the classifiers are trained serially, with the weights on the training instances adaptively set according to the performance of previous classifiers. Our parallel boosting algorithm is designed for tightly coupled shared memory systems with a small number of processors, with an objective of achieving the maximal prediction accuracy in fewer iterations than boosting on a single processor. After all processors learn classifiers in parallel at each boosting round, they are combined according to the confidence of their prediction. Our distributed boosting algorithm is proposed primarily for learning from several disjoint data sites when the data cannot be merged together, although it can also be used for parallel learning where a massive data set is partitioned into several disjoint subsets for a more efficient analysis. At each boosting round, the proposed method combines classifiers from all sites and creates a classifier ensemble on each site. The final classifier is constructed as an ensemble of all classifier ensembles built on disjoint data sets. The new proposed methods applied to several data sets have shown that parallel boosting can achieve the same or even better prediction accuracy considerably faster than the standard sequential boosting. Results from the experiments also indicate that distributed boosting has comparable or slightly improved classification accuracy over standard boosting, while requiring much less memory and computational time since it uses smaller data sets.     

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.     

Constructing ensembles of classifiers using supervised projection methods based on misclassified instances

In this paper, we propose an approach for ensemble construction based on the use of supervised projections, both linear and non-linear, to achieve both accuracy and diversity of individual classifiers. The proposed approach uses the philosophy of boosting, putting more effort on difficult instances, but instead of learning the classifier on a biased distribution of the training set, it uses misclassified instances to find a supervised projection that favors their correct classification. We show that supervised projection algorithms can be used for this task. We try several known supervised projections, both linear and non-linear, in order to test their ability in the present framework. Additionally, the method is further improved introducing concepts from oversampling for imbalance datasets. The introduced method counteracts the negative effect of a low number of instances for constructing the supervised projections.The method is compared with AdaBoost showing an improved performance on a large set of 45 problems from the UCI Machine Learning Repository. Also, the method shows better robustness in presence of noise with respect to AdaBoost.     

The robustness of majority voting compared to filtering misclassified instances in supervised classification tasks

Removing or filtering outliers and mislabeled instances prior to training a learning algorithm has been shown to increase classification accuracy, especially in noisy data sets. A popular approach is to remove any instance that is misclassified by a learning algorithm. However, the use of ensemble methods has also been shown to generally increase classification accuracy. In this paper, we extensively examine filtering and ensembling. We examine 9 learning algorithms individually and ensembled together as filtering algorithms as well as the effects of filtering in the 9 chosen learning algorithms on a set of 54 data sets. We compare the filtering results with using a majority voting ensemble. We find that the majority voting ensemble significantly outperforms filtering unless there are high amounts of noise present in the data set. Additionally, for most cases, using an ensemble of learning algorithms for filtering produces a greater increase in classification accuracy than using a single learning algorithm for filtering.     

结合X-means聚类的自适应随机子空间组合分类算法

针对大规模数据的分类准确率低且效率下降的问题,提出一种结合X-means聚类的自适应随机子空间组合分类算法。首先使用X-means聚类方法,保持原有数据结构的同时,把复杂的数据空间自动分解为多个样本子空间进行分治学习;而自适应随机子空间组合分类器,提升了基分类器的差异性并自动确定基分类器数量,提升了组合分类器的鲁棒性及分类准确性。该算法在人工和UCI数据集上进行了测试,并与传统单分类和组合分类算法进行了比较。实验结果表明,对于大规模数据集,该方法具有更好的分类精度和健壮性,并提升了整体算法的效率。     

基于标记与特征依赖最大化的弱标记集成分类

弱标记学习是多标记学习的一个重要分支,近几年已被广泛研究并被应用于多标记样本的缺失标记补全和预测等问题.然而,针对特征集合较大、更容易拥有多个语义标记和出现标记缺失的高维数据问题,现有弱标记学习方法普遍易受这类数据包含的噪声和冗余特征的干扰.为了对高维多标记数据进行准确的分类,提出了一种基于标记与特征依赖最大化的弱标记集成分类方法EnWL.EnWL首先在高维数据的特征空间多次利用近邻传播聚类方法,每次选择聚类中心构成具有代表性的特征子集,降低噪声和冗余特征的干扰;再在每个特征子集上训练一个基于标记与特征依赖最大化的半监督多标记分类器;最后,通过投票集成这些分类器实现多标记分类.在多种高维数据集上的实验结果表明,EnWL在多种评价度量上的预测性能均优于已有相关方法.     

基于堆叠集成的数据流分类*

对数据流分类分析的常用方法是集成学习。为了得到更好的分类效果,给出一种基于堆叠集成的数据流分类分析方法。该方法通过构造一个分类器对基分类器进行集成。实验结果表明,与基于投票或加权投票的集成方法相比,基于堆叠集成方法对概念漂移的快速适应能力以及预测准确率得到了提高。     

基于Boosting算法集成遗传模糊分类器的文本分类

提出一种新颖的基于Boosting模糊分类的文本分类方法。首先采用潜在语义索引(LSI)对文本特征进行选择;然后提出Boosting算法集成模糊分类器学习,在每轮迭代训练过程中,算法通过调整训练样本的分布,利用遗传算法产生分类规则。减少分类规则能够正确分类样本的权值,使得新产生的分类规则重点考虑难于分类的样本。实验结果表明,该文本分类算法具有良好分类的性能。     

On Mining Instance-Centric Classification Rules

Many studies have shown that rule-based classifiers perform well in classifying categorical and sparse high-dimensional databases. However, a fundamental limitation with many rule-based classifiers is that they find the rules by employing various heuristic methods to prune the search space and select the rules based on the sequential database covering paradigm. As a result, the final set of rules that they use may not be the globally best rules for some instances in the training database. To make matters worse, these algorithms fail to fully exploit some more effective search space pruning methods in order to scale to large databases. In this paper, we present a new classifier, HARMONY, which directly mines the final set of classification rules. HARMONY uses an instance-centric rule-generation approach and it can assure that, for each training instance, one of the highest-confidence rules covering this instance is included in the final rule set, which helps in improving the overall accuracy of the classifier. By introducing several novel search strategies and pruning methods into the rule discovery process, HARMONY also has high efficiency and good scalability. Our thorough performance study with some large text and categorical databases has shown that HARMONY outperforms many well-known classifiers in terms of both accuracy and computational efficiency and scales well with regard to the database size     

VDM-RS: A visual data mining system for exploring and classifying remotely sensed images

Remotely sensed imagery has become increasingly important in several applications domains, such as environmental monitoring, change detection, fire risk mapping and land use, to name only a few. Several advanced image classification techniques have been developed to analyze such imagery and in particular to improve the accuracy of classifying images in the context of such applications. However, most of the proposed classifiers remain a black box to users, leaving them with little to no means to explore and thus further improve the classification process, in particular for misclassified pixel samples. In this paper, we present the concepts, design and implementation of VDM-RS, a visual data mining system for classifying remotely sensed images and exploring image classification processes. The system provides users with two classes of components. First, visual components are offered that are specific to classifying remotely sensed images and provide traditional interfaces, such as a map view and an error matrix view. Second, the decision tree classifier view provides users with the functionality to trace and explore the classification process of individual pixel samples. This feature allows users to inspect how a sample has been correctly classified using the classifier, but more importantly, it also allows for a detailed exploration of the steps in which a sample has been misclassified. The integration of these features into a coherent, user-friendly system not only helps users in getting more insights into the data, but also to better understand and subsequently improve a classifier for remotely sensed images. We demonstrate the functionality of the system's components and their interaction for classifying imagery using a hyperspectral image dataset.     

Automatic prosodic tone choice classification with Brazil’s intonation model

This paper examines the performance of automatically classifying five tone choices (i.e., falling, rising, rising-falling, falling-rising, and neutral) of Brazil’s intonation model. We tested two machine learning classifiers (neural network and boosting ensemble) in two configurations (multi-class and pairwise coupling) and a rule-based classifier. Three sets of acoustic features built from the TILT and Bézier pitch contour models and a new four-point pitch contour model we introduced here were investigated. Tone choices are one of the key elements of Brazil’s prosodic intonation model. We found the rule-based classifier, which was built on our four-point model, achieved better results than the others with an accuracy of 75.1 % and a Cohen’s kappa coefficient of 0.73. This research proves that it is possible to classify tone choices with an accuracy reaching close to the percentage of agreement between two human analysts. The findings further concluded that our four-point model was better for classifying Brazil’s tone choices than both of the TILT or Bézier models.     

A NEW METHOD FOR CONSTRUCTING FUZZY DECISION TREES AND GENERATING FUZZY CLASSIFICATION RULES FROM TRAINING EXAMPLES

This paper presents a new method for constructing fuzzy decision trees and generating fuzzy classification rules from training instances using compound analysis techniques. The proposed method can generate simpler fuzzy classification rules and has a better classification accuracy rate than the existing method. Furthermore, the proposed method generated less fuzzy classification rules.     

基于特征选择的集成分类器抗噪性能分析

特征选择有助于增强集成分类器成员间的随机差异性,从而提高泛化精度。研究了随机子空间法（RandomSub-space）和旋转森林法（RotationForest）两种基于特征选择的集成分类器构造算法,分析讨论了两算法特征选择的方式与随机差异程度之间的关系。通过对UCI数据集引入噪声,比较两者在噪声环境下的分类精度。实验结果表明：当噪声增加及特征关联度下降时,基本学习算法及噪声程度对集成效果均有影响,当噪声增强到一定程度后。集成效果和单分类器的性能趋于一致。     

Information-Based Evaluation Criterion for Classifier's Performance

In the past few years many systems for learning decision rules from examples were developed. As different systems allow different types of answers when classifying new instances, it is difficult to appropriately evaluate the systems' classification power in comparison with other classification systems or in comparison with human experts. Classification accuracy is usually used as a measure of classification performance. This measure is, however, known to have several defects. A fair evaluation criterion should exclude the influence of the class probabilities which may enable a completely uninformed classifier to trivially achieve high classification accuracy. In this paper a method for evaluating the information score of a classifier's answers is proposed. It excludes the influence of prior probabilities, deals with various types of imperfect or probabilistic answers and can be used also for comparing the performance in different domains.