Maximum margin Bayesian network classifiers

We present a maximum margin parameter learning algorithm for Bayesian network classifiers using a conjugate gradient (CG) method for optimization. In contrast to previous approaches, we maintain the normalization constraints on the parameters of the Bayesian network during optimization, i.e., the probabilistic interpretation of the model is not lost. This enables us to handle missing features in discriminatively optimized Bayesian networks. In experiments, we compare the classification performance of maximum margin parameter learning to conditional likelihood and maximum likelihood learning approaches. Discriminative parameter learning significantly outperforms generative maximum likelihood estimation for naive Bayes and tree augmented naive Bayes structures on all considered data sets. Furthermore, maximizing the margin dominates the conditional likelihood approach in terms of classification performance in most cases. We provide results for a recently proposed maximum margin optimization approach based on convex relaxation. While the classification results are highly similar, our CG-based optimization is computationally up to orders of magnitude faster. Margin-optimized Bayesian network classifiers achieve classification performance comparable to support vector machines (SVMs) using fewer parameters. Moreover, we show that unanticipated missing feature values during classification can be easily processed by discriminatively optimized Bayesian network classifiers, a case where discriminative classifiers usually require mechanisms to complete unknown feature values in the data first.     

Restricted Bayesian classification networks

Bayesian networks are graphical models that describe dependency relationships between variables, and are powerful tools for studying probability classifiers. At present, the causal Bayesian network learning method is used in constructing Bayesian network classifiers while the contribution of attribute to class is over-looked. In this paper, a Bayesian network specifically for classification-restricted Bayesian classification networks is proposed. Combining dependency analysis between variables, classification accuracy evaluation criteria and a search algorithm, a learning method for restricted Bayesian classification networks is presented. Experiments and analysis are done using data sets from UCI machine learning repository. The results show that the restricted Bayesian classification network is more accurate than other well-known classifiers.     

基于条件对数似然函数导数的贝叶斯网络分类器优化算法

通常基于鉴别式学习策略训练的贝叶斯网络分类器有较高的精度,但在具有冗余边的网络结构之上鉴别式参数学习算法的性能受到一定的限制.为了在实际应用中进一步提高贝叶斯网络分类器的分类精度,该文定量描述了网络结构与真实数据变量分布之间的关系,提出了一种不存在冗余边的森林型贝叶斯网络分类器及其相应的FAN学习算法(Forest-Augmented Naive Bayes Algorithm),FAN算法能够利用对数条件似然函数的偏导数来优化网络结构学习.实验结果表明常用的限制性贝叶斯网络分类器通常存在一些冗余边,其往往会降低鉴别式参数学习算法的性能;森林型贝叶斯网络分类器减少了结构中的冗余边,更加适合于采用鉴别式学习策略训练参数;应用条件对数似然函数偏导数的FAN算法在大多数实验数据集合上提高了分类精度.     

Enhancing multi-label classification by modeling dependencies among labels

In this paper, we propose a novel framework for multi-label classification, which directly models the dependencies among labels using a Bayesian network. Each node of the Bayesian network represents a label, and the links and conditional probabilities capture the probabilistic dependencies among multiple labels. We employ our Bayesian network structure learning method, which guarantees to find the global optimum structure, independent of the initial structure. After structure learning, maximum likelihood estimation is used to learn the conditional probabilities among nodes. Any current multi-label classifier can be employed to obtain the measurements of labels. Then, using the learned Bayesian network, the true labels are inferred by combining the relationship among labels with the labels? estimates obtained from a current multi-labeling method. We further extend the proposed multi-label classification method to deal with incomplete label assignments. Structural Expectation-Maximization algorithm is adopted for both structure and parameter learning. Experimental results on two benchmark multi-label databases show that our approach can effectively capture the co-occurrent and the mutual exclusive relation among labels. The relation modeled by our approach is more flexible than the pairwise or fixed subset labels captured by current multi-label learning methods. Thus, our approach improves the performance over current multi-label classifiers. Furthermore, our approach demonstrates its robustness to incomplete multi-label classification.     

属性加权的朴素贝叶斯集成分类器

为提高朴素贝叶斯分类器的分类精度和泛化能力,提出了基于属性相关性的加权贝叶斯集成方法（WEBNC）。根据每个条件属性与决策属性的相关度对其赋以相应的权值,然后用AdaBoost训练属性加权后的BNC。该分类方法在16个UCI标准数据集上进行了测试,并与BNC、贝叶斯网和由AdaBoost训练出的BNC进行比较,实验结果表明,该分类器具有更高的分类精度与泛化能力。     

Semisupervised learning of classifiers: theory, algorithms, and their application to human-computer interaction

Automatic classification is one of the basic tasks required in any pattern recognition and human computer interaction application. In this paper, we discuss training probabilistic classifiers with labeled and unlabeled data. We provide a new analysis that shows under what conditions unlabeled data can be used in learning to improve classification performance. We also show that, if the conditions are violated, using unlabeled data can be detrimental to classification performance. We discuss the implications of this analysis to a specific type of probabilistic classifiers, Bayesian networks, and propose a new structure learning algorithm that can utilize unlabeled data to improve classification. Finally, we show how the resulting algorithms are successfully employed in two applications related to human-computer interaction and pattern recognition: facial expression recognition and face detection.     

Structural Extension to Logistic Regression: Discriminative Parameter Learning of Belief Net Classifiers

Bayesian belief nets (BN_s) are often used for classification tasks—typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to find the BN that maximizes a different objective function—viz., likelihood, rather than classification accuracy—typically by first learning an appropriate graphical structure, then finding the parameters for that structure that maximize the likelihood of the data. As these parameters may not maximize the classification accuracy, “discriminative parameter learners” follow the alternative approach of seeking the parameters that maximize conditional likelihood (CL), over the distribution of instances the BN will have to classify. This paper first formally specifies this task, shows how it extends standard logistic regression, and analyzes its inherent sample and computational complexity. We then present a general algorithm for this task, ELR, that applies to arbitrary BN structures and that works effectively even when given incomplete training data. Unfortunately, ELR is not guaranteed to find the parameters that optimize conditional likelihood; moreover, even the optimal-CL parameters need not have minimal classification error. This paper therefore presents empirical evidence that ELR produces effective classifiers, often superior to the ones produced by the standard “generative” algorithms, especially in common situations where the given BN-structure is incorrect.     

基于贝叶斯方法的分类预测

介绍了基本的贝叶斯分类模型和贝叶斯信念网络模型,对网络模型的学习进行了讨论。并从实际出发,提出了几种可以简化模型结构、降低学习复杂性的可行方法,简要说明了这些方法在网络模型中的应用。对贝叶斯分类模型的准确性及其主要特点进行了分析。     

Enhancing network intrusion detection classifiers using supervised adversarial training

The performance of classifiers has a direct impact on the effectiveness of intrusion detection system. Thus, most researchers aim to improve the detection performance of classifiers. However, classifiers can only get limited useful information from the limited number of labeled training samples, which usually affects the generalization of classifiers. In order to enhance the network intrusion detection classifiers, we resort to adversarial training, and a novel supervised learning framework using generative adversarial network for improving the performance of the classifier is proposed in this paper. The generative model in our framework is utilized to continuously generate other complementary labeled samples for adversarial training and assist the classifier for classification, while the classifier in our framework is used to identify different categories. Meanwhile, the loss function is deduced again, and several empirical training strategies are proposed to improve the stabilization of the supervised learning framework. Experimental results prove that the classifier via adversarial training improves the performance indicators of intrusion detection. The proposed framework provides a feasible method to enhance the performance and generalization of the classifier.

     

Aggregative model-based classifier ensemble for improving land-use/cover classification of Landsat TM Images

This article proposes a new approach to improve the classification performance of remotely sensed images with an aggregative model based on classifier ensemble (AMCE). AMCE is a multi-classifier system with two procedures, namely ensemble learning and predictions combination. Two ensemble algorithms (Bagging and AdaBoost.M1) were used in the ensemble learning process to stabilize and improve the performance of single classifiers (i.e. maximum likelihood classifier, minimum distance classifier, back propagation neural network, classification and regression tree, and support vector machine (SVM)). Prediction results from single classifiers were integrated according to a diversity measurement with an averaged double-fault indicator and different combination strategies (i.e. weighted vote, Bayesian product, logarithmic consensus, and behaviour knowledge space). The suitability of the AMCE model was examined using a Landsat Thematic Mapper (TM) image of Dongguan city (Guangdong, China), acquired on 2 January 2009. Experimental results show that the proposed model was significantly better than the most accurate single classification (i.e. SVM) in terms of classification accuracy (i.e. from 88.83% to 92.45%) and kappa coefficient (i.e. from 0.8624 to 0.9088). A stepwise comparison illustrates that both ensemble learning and predictions combination with the AMCE model improved classification.     

Effects of classifier structures and training regimes on integrated segmentation and recognition of handwritten numeral strings

In integrated segmentation and recognition of character strings, the underlying classifier is trained to be resistant to noncharacters. We evaluate the performance of state-of-the-art pattern classifiers of this kind. First, we build a baseline numeral string recognition system with simple but effective presegmentation. The classification scores of the candidate patterns generated by presegmentation are combined to evaluate the segmentation paths and the optimal path is found using the beam search strategy. Three neural classifiers, two discriminative density models, and two support vector classifiers are evaluated. Each classifier has some variations depending on the training strategy: maximum likelihood, discriminative learning both with and without noncharacter samples. The string recognition performances are evaluated on the numeral string images of the NIST special database 19 and the zipcode images of the CEDAR CDROM-1. The results show that noncharacter training is crucial for neural classifiers and support vector classifiers, whereas, for the discriminative density models, the regularization of parameters is important. The string recognition results compare favorably to the best ones reported in the literature though we totally ignored the geometric context. The best results were obtained using a support vector classifier, but the neural classifiers and discriminative density models show better trade-off between accuracy and computational overhead.     

一种基于多进化神经网络的分类方法

分类问题是目前数据挖掘和机器学习领域的重要内容.提出了一种基于多进化神经网络的分类方法CABEN(classification approach based on evolutionary neural networks).利用改进的进化策略和Levenberg-Marquardt方法对多个三层前馈神经网络同时进行训练.训练好各个分类模型以后,将待识别数据分别输入,最后根据绝对多数投票法决定最终分类结果.实验结果表明,该方法可以较好地进行数据分类,而且与传统的神经网络方法以及贝叶斯方法和决策树方法相比,在     

Boosting a fast neural network for supervised land cover classification

It is demonstrated that the use of an ensemble of neural networks for routine land cover classification of multispectral satellite data can lead to a significant improvement in classification accuracy. Specifically, the AdaBoost.M1 algorithm is applied to a sequence of three-layer, feed-forward neural networks. In order to overcome the drawback of long training time for each network in the ensemble, the networks are trained with an efficient Kalman filter algorithm. On the basis of statistical hypothesis tests, classification performance on multispectral imagery is compared with that of maximum likelihood and support vector machine classifiers. Good generalization accuracies are obtained with computation times of the order of 1 h or less. The algorithms involved are described in detail and a software implementation in the ENVI/IDL image analysis environment is provided.     

A flexible probabilistic framework for large-margin mixture of experts

Mixture-of-Experts (MoE) enable learning highly nonlinear models by combining simple expert models. Each expert handles a small region of the data space, as dictated by the gating network which generates the (soft) assignment of input to the corresponding experts. Despite their flexibility and renewed interest lately, existing MoE constructions pose several difficulties during model training. Crucially, neither of the two popular gating networks used in MoE, namely the softmax gating network and hierarchical gating network (the latter used in the hierarchical mixture of experts), have efficient inference algorithms. The problem is further exacerbated if the experts do not have conjugate likelihood and lack a naturally probabilistic formulation (e.g., logistic regression or large-margin classifiers such as SVM). To address these issues, we develop novel inference algorithms with closed-form parameter updates, leveraging some of the recent advances in data augmentation techniques. We also present a novel probabilistic framework for MoE, consisting of a range of gating networks with efficient inference made possible through our proposed algorithms. We exploit this framework by using Bayesian linear SVMs as experts on various classification problems (which has a non-conjugate likelihood otherwise generally), providing our final model with attractive large-margin properties. We show that our models are significantly more efficient than other training algorithms for MoE while outperforming other traditional non-linear models like Kernel SVMs and Gaussian Processes on several benchmark datasets.

     

Using GMDH-based networks for improved spam detection and email feature analysis

Unsolicited or spam email has recently become a major threat that can negatively impact the usability of electronic mail. Spam substantially wastes time and money for business users and network administrators, consumes network bandwidth and storage space, and slows down email servers. In addition, it provides a medium for distributing harmful code and/or offensive content. In this paper, we explore the application of the GMDH (Group Method of Data Handling) based inductive learning approach in detecting spam messages by automatically identifying content features that effectively distinguish spam from legitimate emails. We study the performance for various network model complexities using spambase, a publicly available benchmark dataset. Results reveal that classification accuracies of 91.7% can be achieved using only 10 out of the available 57 attributes, selected through abductive learning as the most effective feature subset (i.e. 82.5% data reduction). We also show how to improve classification performance using abductive network ensembles (committees) trained on different subsets of the training data. Comparison with other techniques such as neural networks and naïve Bayesian classifiers shows that the GMDH-based learning approach can provide better spam detection accuracy with false-positive rates as low as 4.3% and yet requires shorter training time.     

贝叶斯网络分类器近似学习算法

贝叶斯网络在很多领域应用广泛,作为分类器更是一种有效的常用分类方法,它有着很高复杂度,这使得贝叶斯网络分类器在应用中受到诸多限制。通过对贝叶斯网络分类器算法的近似处理,可以有效减少计算量,并且得到令人满意的分类准确率。通过分析一种将判别式算法变为产生式算法的近似方法,介绍了这种算法的近似过程,并将其应用在了贝叶斯网分类算法中。接着对该算法进行分析,利用该算法的稳定性特点,提出Bagging-aCLL 集成分类算法,它进一步提高了该近似算法的分类精度。最后通过实验确定了该算法在分类准确率上确有不错的表现。     

贝叶斯分类器的判别式参数学习

为了提高贝叶斯分类器的分类性能,针对贝叶斯网络分类器的构成特征,提出一种基于参数集成的贝叶斯分类器判别式参数学习算法PEBNC。该算法将贝叶斯分类器的参数学习视为回归问题,将加法回归模型应用于贝叶斯网络分类器的参数学习,实现贝叶斯分类器的判别式参数学习。实验结果表明,在大多数实验数据上,PEBNC能够明显提高贝叶斯分类器的分类准确率。此外,与一般的贝叶斯集成分类器相比,PEBNC不必存储成员分类器的参数,空间复杂度大大降低。     

基于对抗学习的蒙汉神经机器翻译

在机器翻译模型的构建和训练阶段,为了缓解因端到端机器翻译框架在训练时采用最大似然估计原理导致的翻译模型的质量不高的问题,本文使用对抗学习策略训练生成对抗网络,通过鉴别器协助生成器的方式来提高生成器的翻译质量,通过实验选择出了更适合生成器的机器翻译框架Transformer,更适合鉴别器的卷积神经网络,并且验证了对抗式训练对提高译文的自然度、流利度以及准确性都具有一定的作用.在模型的优化阶段,为了缓解因蒙汉平行数据集匮乏导致的蒙汉机器翻译质量仍然不理想的问题,本文将Dual-GAN (dual-generative adversarial networks,对偶生成对抗网络)算法引入了蒙汉机器翻译中,通过有效的利用大量蒙汉单语数据使用对偶学习策略的方式来进一步提高基于对抗学习的蒙汉机器翻译模型的质量.     

基于捕食逃逸PSO的贝叶斯网络分类器

构造精确的贝叶斯网络分类器已被证明为NP难问题,提出了一种基于捕食逃逸粒子群优化(PSO)算法的通用贝叶斯网络分类器,能有效避免数据预处理时的属性约简对分类效果的直接影响,实现对贝叶斯网络结构的精确学习和搜索。另外,将所提出的分类器应用于高职院校就业预测分析,并在Weka平台上实现对该分类器的构建和验证,与其他几种贝叶斯网络分类器的对比实验结果表明,该分类器具有更好的性能。     

Efficient parameter learning of Bayesian network classifiers

Recent advances have demonstrated substantial benefits from learning with both generative and discriminative parameters. On the one hand, generative approaches address the estimation of the parameters of the joint distribution—\(\mathrm{P}(y,\mathbf{x})\), which for most network types is very computationally efficient (a notable exception to this are Markov networks) and on the other hand, discriminative approaches address the estimation of the parameters of the posterior distribution—and, are more effective for classification, since they fit \(\mathrm{P}(y|\mathbf{x})\) directly. However, discriminative approaches are less computationally efficient as the normalization factor in the conditional log-likelihood precludes the derivation of closed-form estimation of parameters. This paper introduces a new discriminative parameter learning method for Bayesian network classifiers that combines in an elegant fashion parameters learned using both generative and discriminative methods. The proposed method is discriminative in nature, but uses estimates of generative probabilities to speed-up the optimization process. A second contribution is to propose a simple framework to characterize the parameter learning task for Bayesian network classifiers. We conduct an extensive set of experiments on 72 standard datasets and demonstrate that our proposed discriminative parameterization provides an efficient alternative to other state-of-the-art parameterizations.