Multi-view Ensemble Research Articles

Improving multi-view ensemble learning with Round-Robin feature set partitioning

Multi-view Ensemble Learning (MEL) techniques have shown remarkable success in improving the accuracy and resilience of classification algorithms by combining multiple base classifiers trained over different perspectives of a dataset, known as views. One crucial factor affecting ensemble performance is the selection of diverse and informative feature subsets. Feature Set Partitioning (FSP) methods address this challenge by creating distinct views of features for each base classifier. In this context, we propose the Round-Robin Feature Set Partitioning (RR-FSP) technique, which introduces a novel approach to feature allocation among views. This novel approach evenly distributes highly correlated features across views, thereby enhancing ensemble diversity, promoting balanced feature utilization, and encouraging the more equitable distribution of correlated features, RR-FSP contributes to the advancement of MEL techniques. Through experiments on various datasets, we demonstrate that RR-FSP offers improved classification accuracy and robustness, making it a valuable addition to the arsenal of FSP techniques for MEL.

Temporomandibular joint CBCT image segmentation via multi-view ensemble learning network.

Accurate segmentation of the temporomandibular joint (TMJ) from cone beam CT (CBCT) images holds significant clinical value for diagnosing temporomandibular joint osteoarthrosis (TMJOA) and related conditions. Convolutional neural network-based medical image segmentation methods have achieved state-of-the-art performance in various segmentation tasks. However, 3D medical images segmentation requires substantial global context and rich spatial semantic information, demanding much more GPU memory and computational resources. To address these challenges in 3D medical image segmentation, we propose a novel network- the MVEL-Net (Multi-view Ensemble Learning Network) for TMJ CBCT image segmentation. By resampling images along three dimensions, we generate multiple weak learners with different spatial semantic information. A subsequent strong learning network effectively integrates the outputs from these weak learners to achieve more accurate segmentation results. We evaluated our network model using a clinical dataset comprising 88 subjects with TMJ CBCT images. The average Dice similarity coefficient (DSC) was 0.9817 ± 0.0049, the average surface distance was 0.0540 ± 0.0179mm, and the 95% Hausdorff distance was 0.1743 ± 0.0550mm. Our proposed MVEL-Net demonstrates excellent segmentation performance on TMJ from CBCT images, while using fewer GPU memory resources compared to other 3D networks. The effectiveness of this method in capturing spatial context could be leveraged for tasks like organ segmentation from volumetric scans. This may facilitate wider adoption of AI-based solutions for automated analysis of 3D medical images.

Semi-supervised Multi-view Ensemble Learning with Consensus

This paper is devoted to enchasing existing multi-view semi-supervised ensemble learning algorithms by introducing a cross-view consensus. A detailed overview of three state-of-the-art methods is given, with relevant steps of the training highlighted. A problem statement is formed to introduce both semi-supervised framework and consider the semi-supervised learning in the context of optimization problem. A novel multi-view semi-supervised ensemble learning algorithm called multi-view semi-supervised cross consensus (MSSXC) is introduced. The algorithm is tested against 5 synthetic datasets designed for semi-supervised learning challenges. The results indicate improvement in the average accuracy of up to 10% in comparison to existing methods, especially in low-volume, high density scenarios.

A multi-view ensemble machine learning approach for 3D modeling using geological and geophysical data

Geophysical data are often integrated into geological data for 3D modeling of underground spaces. However, the existing single-view approach means it is difficult to adequately fuse the valid information between the two types of data, and the complexity of lithological decoding and classification is high. To address this issue, a multi-view ensemble machine learning (ML) framework is proposed. Initially, the original dataset of lithology prediction is constructed by aligning geological and geophysical data with different spatial scales. Next, the dataset is divided into three datasets of structural strength, density, and moisture content according to the lithology properties of the geophysical data. The proposed framework is then used to capture the lithologic characteristics under different views to achieve the prediction of lithologic labels. In this process, a self-attentive mechanism is used to adaptively fuse the valid information under each view. To validate the proposed framework, it is applied to a project in Jiaxing, Zhejiang Province, China. Compared with existing ML methods, the proposed multi-view ensemble ML framework improves modeling accuracy and constructs models with low uncertainty. The framework can be extended to other multi-source data fusion tasks across geoscience domains.

Multiview ensemble clustering of hypergraph p-Laplacian regularization with weighting and denoising

Multiview clustering has gained attention for its ability to incorporate complementary information from multiple sources of data, leading to better clustering results. However, these methods don't sufficiently mine high-dimensional information of non-linear subspace or ignore important high-level information between basic partitions (BPs), which are obtained via single-view clustering that could help to overcome differences between heterogeneous feature spaces and defined in Eq. (4). Additionally, existing multiview ensemble clustering methods neglect the noise arising from the generation phase. In light of this, we first propose a novel multiview subspace clustering method with hypergraph p-Laplacian regularization and denoising. Specifically, to utilize the high-dimensional information, a hypergraph p-Laplacian regularized term is added to the model along with low-rank subspace learning to capture the different hierarchical structures in the data. A denoising algorithm based on KMeans and K-nearest neighbor is used to minimize the noise of similarity matrix. Then utilizing this approach as a backbone, a multiview ensemble clustering of hypergraph p-Laplacian regularization with weighting and denoising method is proposed. A hybrid strategy and a novel global weighting ensemble strategy is further proposed to extract high-level information across all the BPs. By integrating hypergraph p-Laplacian operator, low-rank subspace learning, denoising, and weighting ensemble strategy in a unified framework, this ensemble approach adequately learns latent structures and complementary information. Experimental results on multiple datasets demonstrate the efficacy of this approach.

A multi-view ensemble clustering approach using joint entropy

Multi-view clustering can capture complementary and consistent information from different views, which is a core research topic in the fields of machine learning and pattern recognition. However, existing multi-view clustering methods ignore the quality of each base clustering cluster, which may affect the clustering performance. In this study, a multi-view ensemble clustering approach is proposed using joint entropy to evaluate the base clustering clusters. Firstly, an uncertainty index of base clustering clusters is defined using joint entropy which characterizes the importance and quality of each cluster. Secondly, a weighted co-association matrix is constructed using the uncertainty index, and useless entries are removed from the matrix, making the co-association matrix more reasonable. Thirdly, a candidate clusters selection strategy based on the stability index and a Multi-view Ensemble Clustering (MvEC-DoS) algorithm is proposed. In the end, experiments on five benchmark datasets validate the efficacy of our approach compared to other state-of-the-art methods.

Supervised Parametric Learning in the Identification of Composite Biomarker Signatures of Type 1 Diabetes in Integrated Parallel Multi-Omics Datasets.

Type 1 diabetes (T1D) is a devastating autoimmune disease, and its rising prevalence in the United States and around the world presents a critical problem in public health. While some treatment options exist for patients already diagnosed, individuals considered at risk for developing T1D and who are still in the early stages of their disease pathogenesis without symptoms have no options for any preventive intervention. This is because of the uncertainty in determining their risk level and in predicting with high confidence who will progress, or not, to clinical diagnosis. Biomarkers that assess one's risk with high certainty could address this problem and will inform decisions on early intervention, especially in children where the burden of justifying treatment is high. Single omics approaches (e.g., genomics, proteomics, metabolomics, etc.) have been applied to identify T1D biomarkers based on specific disturbances in association with the disease. However, reliable early biomarkers of T1D have remained elusive to date. To overcome this, we previously showed that parallel multi-omics provides a more comprehensive picture of the disease-associated disturbances and facilitates the identification of candidate T1D biomarkers. This paper evaluated the use of machine learning (ML) using data augmentation and supervised ML methods for the purpose of improving the identification of salient patterns in the data and the ultimate extraction of novel biomarker candidates in integrated parallel multi-omics datasets from a limited number of samples. We also examined different stages of data integration (early, intermediate, and late) to assess at which stage supervised parametric models can learn under conditions of high dimensionality and variation in feature counts across different omics. In the late integration scheme, we employed a multi-view ensemble comprising individual parametric models trained over single omics to address the computational challenges posed by the high dimensionality and variation in feature counts across the different yet integrated multi-omics datasets. the multi-view ensemble improves the prediction of case vs. control and finds the most success in flagging a larger consistent set of associated features when compared with chance models, which may eventually be used downstream in identifying a novel composite biomarker signature of T1D risk. the current work demonstrates the utility of supervised ML in exploring integrated parallel multi-omics data in the ongoing quest for early T1D biomarkers, reinforcing the hope for identifying novel composite biomarker signatures of T1D risk via ML and ultimately informing early treatment decisions in the face of the escalating global incidence of this debilitating disease.

Multi-View Ensemble Federated Learning for Efficient Prediction of Consumer Electronics Applications in Fog Networks

Multi-View Ensemble Federated Learning for Efficient Prediction of Consumer Electronics Applications in Fog Networks

Fossil image identification using deep learning ensembles of data augmented multiviews

Abstract Identification of fossil species is crucial to evolutionary studies. Recent advances from deep learning have shown promising prospects in fossil image identification. However, the quantity and quality of labelled fossil images are often limited due to fossil preservation, conditioned sampling and expensive and inconsistent label annotation by domain experts, which pose great challenges to training deep learning‐based image classification models. To address these challenges, we follow the idea of the wisdom of crowds and propose a multiview ensemble framework, which collects Original (O), Grey (G) and Skeleton (S) views of each fossil image reflecting its different characteristics to train multiple base models, and then makes the final decision via soft voting. Experiments on the largest fusulinid dataset with 2400 images show that the proposed OGS consistently outperforms baselines (using a single model for each view), and obtains superior or comparable performance compared to OOO (using three base models for three the same Original views). Besides, as the training data decreases, the proposed framework achieves more gains. While considering the identification consistency estimation with respect to human experts, OGS receives the highest agreement with the original labels of dataset and with the re‐identifications of two human experts. The validation performance provides a quantitative estimation of consistency across different experts and genera. We conclude that the proposed framework can present state‐of‐the‐art performance in the fusulinid fossil identification case study. This framework is designed for general fossil identification and it is expected to see applications to other fossil datasets in future work. Notably, the result, which shows more performance gains as train set size decreases or over a smaller imbalance fossil dataset, suggests the potential application to identify rare fossil images. The proposed framework also demonstrates its potential for assessing and resolving inconsistencies in fossil identification.

Double High-Order Correlation Preserved Robust Multi-View Ensemble Clustering

Ensemble clustering (EC), utilizing multiple basic partitions (BPs) to yield a robust consensus clustering, has shown promising clustering performance. Nevertheless, most current algorithms suffer from two challenging hurdles: (1) a surge of EC-based methods only focus on pair-wise sample correlation while fully ignoring the high-order correlations of diverse views. (2) they deal directly with the co-association (CA) matrices generated from BPs, which are inevitably corrupted by noise and thus degrade the clustering performance. To address these issues, we propose a novel Double High-Order Correlation Preserved Robust Multi-View Ensemble Clustering (DC-RMEC) method, which preserves the high-order inter-view correlation and the high-order correlation of original data simultaneously. Specifically, DC-RMEC constructs a hypergraph from BPs to fuse high-level complementary information from different algorithms and incorporates multiple CA-based representations into a low-rank tensor to discover the high-order relevance underlying CA matrices, such that double high-order correlation of multi-view features could be dexterously uncovered. Moreover, a marginalized denoiser is invoked to gain robust view-specific CA matrices. Furthermore, we develop a unified framework to jointly optimize the representation tensor and the result matrix. An effective iterative optimization algorithm is designed to optimize our DC-RMEC model by resorting to the alternating direction method of multipliers. Extensive experiments on seven real-world multi-view datasets have demonstrated the superiority of DC-RMEC compared with several state-of-the-art multi-view ensemble clustering methods.

Multi-view cost-sensitive kernel learning for imbalanced classification problem

Multi-view imbalanced learning concentrates on recognizing valuable patterns from multi-view imbalanced data. There are numerous algorithm-level multi-view imbalanced learning methods including multi-view ensemble learning methods and multi-view cost-sensitive learning methods. However, these approaches have two drawbacks. Firstly, from the multi-view representation perspective, they either ignore the agreement among different views, or fail to fully exploit the complementarity information contained in multi-view data. Secondly, from the class-imbalanced perspective, multi-view ensemble learning first needs to design a specific data pre-processing scheme to generate views and then fuses the local and global information with multi-view ensemble schemes. Such models are closely related to practical problems and lack universality. In contrast, multi-view cost-sensitive learning methods possess concise formulations. But the success of such methods depends on imposing appropriate misclassification cost to the majority and minority classes. Inspired by elegant merits of asymmetric LINEX loss function, we propose a multi-view cost-sensitive kernel learning method with LINEX loss termed as MVCSKL. It not only fully exploits the consensus and complementary information of multiple views, but also adaptively learns and adjusts the misclassification cost of different classes through the asymmetric LINEX loss function. Besides, the introduction of the kernel function enables the MVCSKL model to handle complex nonlinear problem and further guarantees its prediction performance. Furthermore, we adopt the alternating direction method of multipliers and gradient descent algorithm to solve MVCSKL. Based on the Rademacher complexity, we analyze the generalization capability of MVCSKL. Extensive experiments are conducted to verify the effectiveness of the proposed model.

Multi-view Ensemble Clustering via Low-rank and Sparse Decomposition: From Matrix to Tensor

As a significant extension of classical clustering methods, ensemble clustering first generates multiple basic clusterings and then fuses them into one consensus partition by solving a problem concerning graph partition with respect to the co-association matrix. Although the collaborative cluster structure among basic clusterings can be well discovered by ensemble clustering, most advanced ensemble clustering utilizes the self-representation strategy with the constraint of low-rank to explore a shared consensus representation matrix in multiple views. However, they still encounter two challenges: (1) high computational cost caused by both the matrix inversion operation and singular value decomposition of large-scale square matrices; (2) less considerable attention on high-order correlation attributed to the pursue of the two-dimensional pair-wise relationship matrix. In this article, based on low-rank and sparse decomposition from both matrix and tensor perspectives, we propose two novel multi-view ensemble clustering methods, which tangibly decrease computational complexity. Specifically, our first method utilizes low-rank and sparse matrix decomposition to learn one common co-association matrix, while our last method constructs all co-association matrices into one third-order tensor to investigate the high-order correlation among multiple views by low-rank and sparse tensor decomposition. We adopt the alternating direction method of multipliers to solve two convex models by dividing them into several subproblems with closed-form solution. Experimental results on ten real-world datasets prove the effectiveness and efficiency of the proposed two multi-view ensemble clustering methods by comparing them with other advanced ensemble clustering methods.

Brain Tumor Segmentation Network with Multi-View Ensemble Discrimination and Kernel-Sharing Dilated Convolution

Accurate segmentation of brain tumors from magnetic resonance 3D images (MRI) is critical for clinical decisions and surgical planning. Radiologists usually separate and analyze brain tumors by combining images of axial, coronal, and sagittal views. However, traditional convolutional neural network (CNN) models tend to use information from only a single view or one by one. Moreover, the existing models adopt a multi-branch structure with different-size convolution kernels in parallel to adapt to various tumor sizes. However, the difference in the convolution kernels' parameters cannot precisely characterize the feature similarity of tumor lesion regions with various sizes, connectivity, and convexity. To address the above problems, we propose a hierarchical multi-view convolution method that decouples the standard 3D convolution into axial, coronal, and sagittal views to provide complementary-view features. Then, every pixel is classified by ensembling the discriminant results from the three views. Moreover, we propose a multi-branch kernel-sharing mechanism with a dilated rate to obtain parameter-consistent convolution kernels with different receptive fields. We use the BraTS2018 and BraTS2020 datasets for comparison experiments. The average Dice coefficients of the proposed network on the BraTS2020 dataset can reach 78.16%, 89.52%, and 83.05% for the enhancing tumor (ET), whole tumor (WT), and tumor core (TC), respectively, while the number of parameters is only 0.5 M. Compared with the baseline network for brain tumor segmentation, the accuracy was improved by 1.74%, 0.5%, and 2.19%, respectively.

Combining multi-view ensemble and surrogate lagrangian relaxation for real-time 3D biomedical image segmentation on the edge

Real-time 3D biomedical image segmentation is always preferred considering the exponentially growing medical imaging data for the past decade. Recently deep learning has significantly boosted the performance of automatic medical image segmentation with high computation and memory requirements, especially for 3D biomedical images. Meanwhile, the privacy and security of patient data have always been the primary concern in medical applications among hospitals and clinics, and there also exists some applications which need real-time processing in clinic practice. Thus, 3D biomedical image segmentation is typically required to be performed locally (i.e. on the edge) with limited computation and memory resources. In this paper, we propose to combine multi-view ensemble and Surrogate Lagrangian relaxation (SLR) for real-time 3D biomedical image segmentation on the edge. Instead of directly dealing with 3D biomedical images, our segmentation conducts on the three 2D domains of the 3D images with an ensemble strategy. In addition, Surrogate Lagrangian relaxation is proposed to compress the model to enable high efficiency and real-time processing. Experiments on a typical edge Nvidia GPU show that our method achieves real-time processing which is 1.5× faster with an improvement of 9% on accuracy compared with single-view models. It also saves 26× computational resources and 6× memory resources compared to 3D segmentation models.

A Novel Multi-View Ensemble Learning Architecture to Improve the Structured Text Classification

Multi-view ensemble learning exploits the information of data views. To test its efficiency for full text classification, a technique has been implemented where the views correspond to the document sections. For classification and prediction, we use a stacking generalization based on the idea that different learning algorithms provide complementary explanations of the data. The present study implements the stacking approach using support vector machine algorithms as the baseline and a C4.5 implementation as the meta-learner. Views are created with OHSUMED biomedical full text documents. Experimental results lead to the sustained conclusion that the application of multi-view techniques to full texts significantly improves the task of text classification, providing a significant contribution for the biomedical text mining research. We also have evidence to conclude that enriched datasets with text from certain sections are better than using only titles and abstracts.

A Multi-View Ensemble Width-Depth Neural Network for Short-Term Wind Power Forecasting

Short-term wind power forecasting (SWPF) is essential for managing wind power systems management. However, most existing forecasting methods fail to fully consider how to rationally integrate multi-view learning technologies with attention mechanisms. In this case, some potential features cannot be fully extracted, degenerating the predictive accuracy and robustness in SWPF. To solve this problem, this paper proposes a multi-view ensemble width-depth neural network (MVEW-DNN) for SWPF. Specifically, MVEW-DNN consists of local and global view learning subnetworks, which can effectively achieve more potential global and local view features of the original wind power data. In MVEW-DNN, the local view learning subnetwork is developed by introducing the deep belief network (DBN) model, which can efficiently extract the local view features. On the other hand, by introducing the attention mechanism, a new deep encoder board learning system (deBLS) is developed as the global view learning subnetwork, which provides more comprehensive global information. Therefore, by rationally learning the effective local and global view features, MVEW-DNN can achieve competitive predictive performance in SWPF. MVEW-DNN is compared with the state-of-the-art models in SWPF. The experiment results indicate that MVEW-DNN can provide competitive predictive accuracy and robustness.

Automated 2D Slice-Based Skull Stripping Multi-View Ensemble Model on NFBS and IBSR Datasets.

This study proposed and evaluated a two-dimensional (2D) slice-based multi-view U-Net (MVU-Net) architecture for skull stripping. The proposed model fused all three TI-weighted brain magnetic resonance imaging (MRI) views, i.e., axial, coronal, and sagittal. This 2D method performed equally well as a three-dimensional (3D) model of skull stripping. while using fewer computational resources. The predictions of all three views were fused linearly, producing a final brain mask with better accuracy and efficiency. Meanwhile, two publicly available datasets-the Internet Brain Segmentation Repository (IBSR) and Neurofeedback Skull-stripped (NFBS) repository-were trained and tested. The MVU-Net, U-Net, and skip connection U-Net (SCU-Net) architectures were then compared. For the IBSR dataset, compared to U-Net and SC-UNet, the MVU-Net architecture attained better mean dice score coefficient (DSC), sensitivity, and specificity, at 0.9184, 0.9397, and 0.9908, respectively. Similarly, the MVU-Net architecture achieved better mean DSC, sensitivity, and specificity, at 0.9681, 0.9763, and 0.9954, respectively, than the U-Net and SC-UNet for the NFBS dataset.

Multi-View Ensemble Learning for Robust Graph Convolutional Networks

Multi-View Ensemble Learning for Robust Graph Convolutional Networks

Ensemble sparse representation-based cyber threat hunting for security of smart cities

The ever-growing expansion of smart cities and the Internet of Things (IoT) offer a promising solution to many contemporary urban challenges. However, this digital transformation also results in cyber-security loopholes which can be exploited by malicious hackers to wreak substantial digital and physical damage. Malware is the primary tool of cyber-criminals for attacking digital systems. In this paper, a multi-view ensemble threat hunting model based on Sparse Representation based Classifier (SRC) is proposed to use in IoT systems that are finding domain space in the advent of Smart Cities. An ensemble of SRCs is considered where every individual SRC classifies malware by Opcode, Bytecode and system call views of several standard IoT and Ransomware datasets. The final decision is made through weighted majority voting. SRC is employed to alleviate the complexity of the base classifiers. Experimental results verify the efficiency and robustness of the proposed model in different balanced and imbalanced environments. The proposed model outperforms all base classifiers and several well-known works in current literature.

A novel approach for classification of mental tasks using multiview ensemble learning (MEL)

Brain-computer interface (BCI) is a domain, in which a person can send information without using any exterior nerve or muscles, just using their brain signal, called electroencephalography (EEG) signal. Multiview learning or data integration or data fusion from a different set of features is an emerging way in machine learning to improve the generalized performance by considering the knowledge with multiple views. Multiview learning has made rapid progress and development in recent years and is also facing many new challenges. This method can be used in the BCI domain, as the meaningful representation of the EEG signal in plenty of ways. This study utilized the multiview ensemble learning (MEL) approach for the binary classification of five mental tasks on the six subjects individually. In this study, we used a well-known EEG database (Keirn and Aunon database). The EEG signal has been decomposed using by methods i.e wavelet transform (WT), empirical mode decomposition (EMD), empirical wavelet transform (EWT), and fuzzy C-means followed by EWT (FEWT). After that, the feature coding technique is applied using parametric feature formation from the decomposed signal. Hence, we had four views to learn four same type of independent base classifiers and predictions are made in an ensemble manner. The study is performed independently with three types of base classifiers, i.e., K-nearest neighbor (KNN), support vector machine (SVM) with linear and non-linear kernels The performance validation of the ten combinations of mental tasks was performed by three MEL based classifiers, i.e., K-nearest neighbor (KNN), support vector machine (SVM) with linear and non-linear kernels. For reliability of the obtained results of the classifiers, 10-fold cross-validation was used. The proposed algorithm shows a promising accuracy of 80% to 100% for binary pair-wise classification of mental tasks.