Adaptive Semi-Supervised Classifier Ensemble for High Dimensional Data Classification

Semisupervised Manifold Regularization via a Subnetwork-Based Representation Learning Model

Semisupervised classification with a few labeled training samples is a challenging task in the area of data mining. Moore-Penrose inverse (MPI)-based manifold regularization (MR) is a widely used technique in tackling semisupervised classification. However, most of the existing MPI-based MR algorithms can only generate loosely connected feature encoding, which is generally less effective in data representation and feature learning. To alleviate this deficiency, we introduce a new semisupervised multilayer subnet neural network called SS-MSNN. The key contributions of this article are as follows: 1) a novel MPI-based MR model using the subnetwork structure is introduced. The subnet model is utilized to enrich the latent space representations iteratively; 2) a one-step training process to learn the discriminative encoding is proposed. The proposed SS-MSNN learns parameters by directly optimizing the entire network, accepting input from one end, and producing output at the other end; and 3) a new semisupervised dataset called HFSWR-RDE is built for this research. Experimental results on multiple domains show that the SS-MSNN achieves promising performance over the other semisupervised learning algorithms, demonstrating fast inference speed and better generalization ability.

Adaptive Subspace Optimization Ensemble Method for High-Dimensional Imbalanced Data Classification

It is hard to construct an optimal classifier for high-dimensional imbalanced data, on which the performance of classifiers is seriously affected and becomes poor. Although many approaches, such as resampling, cost-sensitive, and ensemble learning methods, have been proposed to deal with the skewed data, they are constrained by high-dimensional data with noise and redundancy. In this study, we propose an adaptive subspace optimization ensemble method (ASOEM) for high-dimensional imbalanced data classification to overcome the above limitations. To construct accurate and diverse base classifiers, a novel adaptive subspace optimization (ASO) method based on adaptive subspace generation (ASG) process and rotated subspace optimization (RSO) process is designed to generate multiple robust and discriminative subspaces. Then a resampling scheme is applied on the optimized subspace to build a class-balanced data for each base classifier. To verify the effectiveness, our ASOEM is implemented based on different resampling strategies on 24 real-world high-dimensional imbalanced datasets. Experimental results demonstrate that our proposed methods outperform other mainstream imbalance learning approaches and classifier ensemble methods.

A Structure Constraint Matrix Factorization Framework for Human Behavior Segmentation

This article presents a structure constraint matrix factorization framework for different behavior segmentation of the human behavior sequential data. This framework is based on the structural information of the behavior continuity and the high similarity between neighboring frames. Due to the high similarity and high dimensionality of human behavior data, the high-precision segmentation of human behavior is hard to achieve from the perspective of application and academia. By making the behavior continuity hypothesis, first, the effective constraint regular terms are constructed. Subsequently, the clustering framework based on constrained non-negative matrix factorization is established. Finally, the segmentation result can be obtained by using the spectral clustering and graph segmentation algorithm. For illustration, the proposed framework is applied to the Weiz dataset, Keck dataset, mo_86 dataset, and mo_86_9 dataset. Empirical experiments on several public human behavior datasets demonstrate that the structure constraint matrix factorization framework can automatically segment human behavior sequences. Compared to the classical algorithm, the proposed framework can ensure consistent segmentation of sequential points within behavior actions and provide better performance in accuracy.

Learning High-Dimensional Evolving Data Streams With Limited Labels

In the context of streaming data, learning algorithms often need to confront several unique challenges, such as concept drift, label scarcity, and high dimensionality. Several concept drift-aware data stream learning algorithms have been proposed to tackle these issues over the past decades. However, most existing algorithms utilize a supervised learning framework and require all true class labels to update their models. Unfortunately, in the streaming environment, requiring all labels is unfeasible and not realistic in many real-world applications. Therefore, learning data streams with minimal labels is a more practical scenario. Considering the problem of the curse of dimensionality and label scarcity, in this article, we present a new semisupervised learning technique for streaming data. To cure the curse of dimensionality, we employ a denoising autoencoder to transform the high-dimensional feature space into a reduced, compact, and more informative feature representation. Furthermore, we use a cluster-and-label technique to reduce the dependency on true class labels. We employ a synchronization-based dynamic clustering technique to summarize the streaming data into a set of dynamic microclusters that are further used for classification. In addition, we employ a disagreement-based learning method to cope with concept drift. Extensive experiments performed on many real-world datasets demonstrate the superior performance of the proposed method compared to several state-of-the-art methods.

Progressive Ensemble Kernel-Based Broad Learning System for Noisy Data Classification

The broad learning system (BLS) is an algorithm that facilitates feature representation learning and data classification. Although weights of BLS are obtained by analytical computation, which brings better generalization and higher efficiency, BLS suffers from two drawbacks: 1) the performance depends on the number of hidden nodes, which requires manual tuning, and 2) double random mappings bring about the uncertainty, which leads to poor resistance to noise data, as well as unpredictable effects on performance. To address these issues, a kernel-based BLS (KBLS) method is proposed by projecting feature nodes obtained from the first random mapping into kernel space. This manipulation reduces the uncertainty, which contributes to performance improvements with the fixed number of hidden nodes, and indicates that manually tuning is no longer needed. Moreover, to further improve the stability and noise resistance of KBLS, a progressive ensemble framework is proposed, in which the residual of the previous base classifiers is used to train the following base classifier. We conduct comparative experiments against the existing state-of-the-art hierarchical learning methods on multiple noisy real-world datasets. The experimental results indicate our approaches achieve the best or at least comparable performance in terms of accuracy.

Two-Stage Selective Ensemble of CNN via Deep Tree Training for Medical Image Classification

Medical image classification is an important task in computer-aided diagnosis systems. Its performance is critically determined by the descriptiveness and discriminative power of features extracted from images. With rapid development of deep learning, deep convolutional neural networks (CNNs) have been widely used to learn the optimal high-level features from the raw pixels of images for a given classification task. However, due to the limited amount of labeled medical images with certain quality distortions, such techniques crucially suffer from the training difficulties, including overfitting, local optimums, and vanishing gradients. To solve these problems, in this article, we propose a two-stage selective ensemble of CNN branches via a novel training strategy called deep tree training (DTT). In our approach, DTT is adopted to jointly train a series of networks constructed from the hidden layers of CNN in a hierarchical manner, leading to the advantage that vanishing gradients can be mitigated by supplementing gradients for hidden layers of CNN, and intrinsically obtain the base classifiers on the middle-level features with minimum computation burden for an ensemble solution. Moreover, the CNN branches as base learners are combined into the optimal classifier via the proposed two-stage selective ensemble approach based on both accuracy and diversity criteria. Extensive experiments on CIFAR-10 benchmark and two specific medical image datasets illustrate that our approach achieves better performance in terms of accuracy, sensitivity, specificity, and F1 score measurement.

Adaptive Dense Ensemble Model for Text Classification

Text classification has been widely explored in natural language processing. In this article, we propose a novel adaptive dense ensemble model (AdaDEM) for text classification, which includes local ensemble stage (LES) and global dense ensemble stage (GDES). To strengthen the classification ability and robustness of the enhanced layer, we propose a selective ensemble model based on enhanced attention convolutional neural networks (EnCNNs). To increase the diversity of the ensemble system, these EnCNNs are generated by using two manners: 1) different sample subsets and 2) different granularity kernels. Then, an evaluation criterion that considers both accuracy and diversity is proposed in LES to obtain effective integration results. Furthermore, to make better use of information flow, we develop an adaptive dense ensemble structure with multiple enhanced layers in GDES to mitigate the issue that there may be redundant or invalid enhanced layers in the cascade structure. We conducted extensive experiments against state-of-the-art methods on multiple real-world datasets, including long and short texts, which has verified the effectiveness and generality of our method.

Cross-Lingual Knowledge Transferring by Structural Correspondence and Space Transfer

The cross-lingual sentiment analysis (CLSA) aims to leverage label-rich resources in the source language to improve the models of a resource-scarce domain in the target language, where monolingual approaches based on machine learning usually suffer from the unavailability of sentiment knowledge. Recently, the transfer learning paradigm that can transfer sentiment knowledge from resource-rich languages, for example, English, to resource-poor languages, for example, Chinese, has gained particular interest. Along this line, in this article, we propose semisupervised learning with SCL and space transfer (ssSCL-ST), a semisupervised transfer learning approach that makes use of structural correspondence learning as well as space transfer for cross-lingual sentiment analysis. The key idea behind ssSCL-ST, at a high level, is to explore the intrinsic sentiment knowledge in the target-lingual domain and to reduce the loss of valuable knowledge due to the knowledge transfer via semisupervised learning. ssSCL-ST also features in pivot set extension and space transfer, which helps to enhance the efficiency of knowledge transfer and improve the classification accuracy in the target language domain. Extensive experimental results demonstrate the superiority of ssSCL-ST to the state-of-the-art approaches without using any parallel corpora.

An Evolutionary Multitasking-Based Feature Selection Method for High-Dimensional Classification

Feature selection (FS) is an important data preprocessing technique in data mining and machine learning, which aims to select a small subset of information features to increase the performance and reduce the dimensionality. Particle swarm optimization (PSO) has been successfully applied to FS due to being efficient and easy to implement. However, most of the existing PSO-based FS methods face the problems of trapping into local optima and computationally expensive high-dimensional data. Multifactorial optimization (MFO), as an effective evolutionary multitasking paradigm, has been widely used for solving complex problems through implicit knowledge transfer between related tasks. Inspired by MFO, this study proposes a novel PSO-based FS method to solve high-dimensional classification via information sharing between two related tasks generated from a dataset. To be specific, two related tasks about the target concept are established by evaluating the importance of features. A new crossover operator, called assortative mating, is applied to share information between these two related tasks. In addition, two mechanisms, which are variable-range strategy and subset updating mechanism, are also developed to reduce the search space and maintain the diversity of the population, respectively. The results show that the proposed FS method can achieve higher classification accuracy with a smaller feature subset in a reasonable time than the state-of-the-art FS methods on the examined high-dimensional classification problems.

Classifier Ensemble Based on Multiview Optimization for High-Dimensional Imbalanced Data Classification

High-dimensional class imbalanced data have plagued the performance of classification algorithms seriously. Because of a large number of redundant/invalid features and the class imbalanced issue, it is difficult to construct an optimal classifier for high-dimensional imbalanced data. Classifier ensemble has attracted intensive attention since it can achieve better performance than an individual classifier. In this work, we propose a multiview optimization (MVO) to learn more effective and robust features from high-dimensional imbalanced data, based on which an accurate and robust ensemble system is designed. Specifically, an optimized subview generation (OSG) in MVO is first proposed to generate multiple optimized subviews from different scenarios, which can strengthen the classification ability of features and increase the diversity of ensemble members simultaneously. Second, a new evaluation criterion that considers the distribution of data in each optimized subview is developed based on which a selective ensemble of optimized subviews (SEOS) is designed to perform the subview selective ensemble. Finally, an oversampling approach is executed on the optimized view to obtain a new class rebalanced subset for the classifier. Experimental results on 25 high-dimensional class imbalanced datasets indicate that the proposed method outperforms other mainstream classifier ensemble methods.

Semisupervised Multiple Choice Learning for Ensemble Classification

Ensemble learning has many successful applications because of its effectiveness in boosting the predictive performance of classification models. In this article, we propose a semisupervised multiple choice learning (SemiMCL) approach to jointly train a network ensemble on partially labeled data. Our model mainly focuses on improving a labeled data assignment among the constituent networks and exploiting unlabeled data to capture domain-specific information, such that semisupervised classification can be effectively facilitated. Different from conventional multiple choice learning models, the constituent networks learn multiple tasks in the training process. Specifically, an auxiliary reconstruction task is included to learn domain-specific representation. For the purpose of performing implicit labeling on reliable unlabeled samples, we adopt a negative l₁ -norm regularization when minimizing the conditional entropy with respect to the posterior probability distribution. Extensive experiments on multiple real-world datasets are conducted to verify the effectiveness and superiority of the proposed SemiMCL model.

Laplacian Welsch Regularization for Robust Semisupervised Learning

Semisupervised learning (SSL) has been widely used in numerous practical applications where the labeled training examples are inadequate while the unlabeled examples are abundant. Due to the scarcity of labeled examples, the performances of the existing SSL methods are often affected by the outliers in the labeled data, leading to the imperfect trained classifier. To enhance the robustness of SSL methods to the outliers, this article proposes a novel SSL algorithm called Laplacian Welsch regularization (LapWR). Specifically, apart from the conventional Laplacian regularizer, we also introduce a bounded, smooth, and nonconvex Welsch loss which can suppress the adverse effect brought by the labeled outliers. To handle the model nonconvexity caused by the Welsch loss, an iterative half-quadratic (HQ) optimization algorithm is adopted in which each subproblem has an ideal closed-form solution. To handle the large datasets, we further propose an accelerated model by utilizing the Nyström method to reduce the computational complexity of LapWR. Theoretically, the generalization bound of LapWR is derived based on analyzing its Rademacher complexity, which suggests that our proposed algorithm is guaranteed to obtain satisfactory performance. By comparing LapWR with the existing representative SSL algorithms on various benchmark and real-world datasets, we experimentally found that LapWR performs robustly to outliers and is able to consistently achieve the top-level results.

RCTE: A reliable and consistent temporal-ensembling framework for semi-supervised segmentation of COVID-19 lesions

The segmentation of COVID-19 lesions from computed tomography (CT) scans is crucial to develop an efficient automated diagnosis system. Deep learning (DL) has shown success in different segmentation tasks. However, an efficient DL approach requires a large amount of accurately annotated data, which is difficult to aggregate owing to the urgent situation of COVID-19. Inaccurate annotation can easily occur without experts, and segmentation performance is substantially worsened by noisy annotations. Therefore, this study presents a reliable and consistent temporal-ensembling (RCTE) framework for semi-supervised lesion segmentation. A segmentation network is integrated into a teacher-student architecture to segment infection regions from a limited number of annotated CT scans and a large number of unannotated CT scans. The network generates reliable and unreliable targets, and to evenly handle these targets potentially degrades performance. To address this, a reliable teacher-student architecture is introduced, where a reliable teacher network is the exponential moving average (EMA) of a reliable student network that is reliably renovated by restraining the student involvement to EMA when its loss is larger. We also present a noise-aware loss based on improvements to generalized cross-entropy loss to lead the segmentation performance toward noisy annotations. Comprehensive analysis validates the robustness of RCTE over recent cutting-edge semi-supervised segmentation techniques, with a 65.87% Dice score.

Evolving Multiobjective Cancer Subtype Diagnosis From Cancer Gene Expression Data

Detection and diagnosis of cancer are especially essential for early prevention and effective treatments. Many studies have been proposed to tackle the subtype diagnosis problems with those data, which often suffer from low diagnostic ability and bad generalization. This article studies a multiobjective PSO-based hybrid algorithm (MOPSOHA) to optimize four objectives including the number of features, the accuracy, and two entropy-based measures: the relevance and the redundancy simultaneously, diagnosing the cancer data with high classification power and robustness. First, we propose a novel binary encoding strategy to choose informative gene subsets to optimize those objective functions. Second, a mutation operator is designed to enhance the exploration capability of the swarm. Finally, a local search method based on the "best/1" mutation operator of differential evolutionary algorithm (DE) is employed to exploit the neighborhood area with sparse high-quality solutions since the base vector always approaches to some good promising areas. In order to demonstrate the effectiveness of MOPSOHA, it is tested on 41 cancer datasets including thirty-five cancer gene expression datasets and six independent disease datasets. Compared MOPSOHA with other state-of-the-art algorithms, the performance of MOPSOHA is superior to other algorithms in most of the benchmark datasets.

End-to-End Ensemble Learning by Exploiting the Correlation Between Individuals and Weights

Ensemble learning performs better than a single classifier in most tasks due to the diversity among multiple classifiers. However, the enhancement of the diversity is at the expense of reducing the accuracies of individual classifiers in general and, thus, how to balance the diversity and accuracies is crucial for improving the ensemble performance. In this paper, we propose a new ensemble method which exploits the correlation between individual classifiers and their corresponding weights by constructing a joint optimization model to achieve the tradeoff between the diversity and the accuracy. Specifically, the proposed framework can be modeled as a shallow network and efficiently trained by the end-to-end manner. In the proposed ensemble method, not only can a high total classification performance be achieved by the weighted classifiers but also the individual classifier can be updated based on the error of the optimized weighted classifiers ensemble. Furthermore, the sparsity constraint is imposed on the weight to enforce that partial individual classifiers are selected for final classification. Finally, the experimental results on the UCI datasets demonstrate that the proposed method effectively improves the performance of classification compared with relevant existing ensemble methods.

Genetic Programming With a New Representation to Automatically Learn Features and Evolve Ensembles for Image Classification

Image classification is a popular task in machine learning and computer vision, but it is very challenging due to high variation crossing images. Using ensemble methods for solving image classification can achieve higher classification performance than using a single classification algorithm. However, to obtain a good ensemble, the component (base) classifiers in an ensemble should be accurate and diverse. To solve image classification effectively, feature extraction is necessary to transform raw pixels into high-level informative features. However, this process often requires domain knowledge. This article proposes an evolutionary approach based on genetic programming to automatically and simultaneously learn informative features and evolve effective ensembles for image classification. The new approach takes raw images as inputs and returns predictions of class labels based on the evolved classifiers. To achieve this, a new individual representation, a new function set, and a new terminal set are developed to allow the new approach to effectively find the best solution. More important, the solutions of the new approach can extract informative features from raw images and can automatically address the diversity issue of the ensembles. In addition, the new approach can automatically select and optimize the parameters for the classification algorithms in the ensemble. The performance of the new approach is examined on 13 different image classification datasets of varying difficulty and compared with a large number of effective methods. The results show that the new approach achieves better classification accuracy on most datasets than the competitive methods. Further analysis demonstrates that the new approach can evolve solutions with high accuracy and diversity.

Pairwise Constraint Propagation With Dual Adversarial Manifold Regularization

Pairwise constraints (PCs) composed of must-links (MLs) and cannot-links (CLs) are widely used in many semisupervised tasks. Due to the limited number of PCs, pairwise constraint propagation (PCP) has been proposed to augment them. However, the existing PCP algorithms only adopt a single matrix to contain all the information, which overlooks the differences between the two types of links such that the discriminability of the propagated PCs is compromised. To this end, this article proposes a novel PCP model via dual adversarial manifold regularization to fully explore the potential of the limited initial PCs. Specifically, we propagate MLs and CLs with two separated variables, called similarity and dissimilarity matrices, under the guidance of the graph structure constructed from data samples. At the same time, the adversarial relationship between the two matrices is taken into consideration. The proposed model is formulated as a nonnegative constrained minimization problem, which can be efficiently solved with convergence theoretically guaranteed. We conduct extensive experiments to evaluate the proposed model, including propagation effectiveness and applications on constrained clustering and metric learning, all of which validate the superior performance of our model to state-of-the-art PCP models.

Deep Network Embedding for Graph Representation Learning in Signed Networks

Network embedding has attracted an increasing attention over the past few years. As an effective approach to solve graph mining problems, network embedding aims to learn a low-dimensional feature vector representation for each node of a given network. The vast majority of existing network embedding algorithms, however, are only designed for unsigned networks, and the signed networks containing both positive and negative links, have pretty distinct properties from the unsigned counterpart. In this paper, we propose a deep network embedding model to learn the low-dimensional node vector representations with structural balance preservation for the signed networks. The model employs a semisupervised stacked auto-encoder to reconstruct the adjacency connections of a given signed network. As the adjacency connections are overwhelmingly positive in the real-world signed networks, we impose a larger penalty to make the auto-encoder focus more on reconstructing the scarce negative links than the abundant positive links. In addition, to preserve the structural balance property of signed networks, we design the pairwise constraints to make the positively connected nodes much closer than the negatively connected nodes in the embedding space. Based on the network representations learned by the proposed model, we conduct link sign prediction and community detection in signed networks. Extensive experimental results in real-world datasets demonstrate the superiority of the proposed model over the state-of-the-art network embedding algorithms for graph representation learning in signed networks.

An Ensemble Model for Multi-Level Speech Emotion Recognition

Speech emotion recognition is a challenging and widely examined research topic in the field of speech processing. The accuracy of existing models in speech emotion recognition tasks is not high, and the generalization ability is not strong. Since the feature set and model design of effective speech directly affect the accuracy of speech emotion recognition, research on features and models is important. Because emotional expression is often correlated with the global features, local features, and model design of speech, it is often difficult to find a universal solution for effective speech emotion recognition. Based on this, the main research purpose of this paper is to generate general emotion features in speech signals from different angles, and use the ensemble learning model to perform emotion recognition tasks. It is divided into the following aspects: (1) Three expert roles of speech emotion recognition are designed. Expert 1 focuses on three-dimensional feature extraction of local signals; expert 2 focuses on extraction of comprehensive information in local data; and expert 3 emphasizes global features: acoustic feature descriptors (low-level descriptors (LLDs)), high-level statistics functionals (HSFs), and local features and their timing relationships. A single-/multiple-level deep learning model that meets expert characteristics is designed for each expert, including convolutional neural network (CNN), bi-directional long short-term memory (BLSTM), and gated recurrent unit (GRU). Convolutional recurrent neural network (CRNN), based on a combination of an attention mechanism, is used for internal training of experts. (2) By designing an ensemble learning model, each expert can play to its own advantages and evaluate speech emotions from different focuses. (3) Through experiments, the performance of various experts and ensemble learning models in emotion recognition is compared in the Interactive Emotional Dyadic Motion Capture (IEMOCAP) corpus and the validity of the proposed model is verified.

Hybrid Incremental Ensemble Learning for Noisy Real-World Data Classification

Traditional ensemble learning approaches explore the feature space and the sample space, respectively, which will prevent them to construct more powerful learning models for noisy real-world dataset classification. The random subspace method only search for the selection of features. Meanwhile, the bagging approach only search for the selection of samples. To overcome these limitations, we propose the hybrid incremental ensemble learning (HIEL) approach which takes into consideration the feature space and the sample space simultaneously to handle noisy dataset. Specifically, HIEL first adopts the bagging technique and linear discriminant analysis to remove noisy attributes, and generates a set of bootstraps and the corresponding ensemble members in the subspaces. Then, the classifiers are selected incrementally based on a classifier-specific criterion function and an ensemble criterion function. The corresponding weights for the classifiers are assigned during the same process. Finally, the final label is summarized by a weighted voting scheme, which serves as the final result of the classification. We also explore various classifier-specific criterion functions based on different newly proposed similarity measures, which will alleviate the effect of noisy samples on the distance functions. In addition, the computational cost of HIEL is analyzed theoretically. A set of nonparametric tests are adopted to compare HIEL and other algorithms over several datasets. The experiment results show that HIEL performs well on the noisy datasets. HIEL outperforms most of the compared classifier ensemble methods on 14 out of 24 noisy real-world UCI and KEEL datasets.