Narrow Results

In recent years, the application of machine learning techniques for identifying epileptic electroencephalogram (EEG) signals has become increasingly prevalent. In this study, we propose a method called Multi-View Learning and Regularized Label Relaxation Linear Regression (MVRLR) for the recognition of epileptic EEG signals. To be specific, first, we generate multi-view data by employing three different feature extraction methods to capture the EEG features from different perspectives. Then, in the linear regression framework, we introduce a nonnegative label relaxation matrix to transform the strict binary label matrix into a relaxed variable matrix, allowing for more flexible label fitting and expanding the boundaries between different classes. Additionally, we utilize manifold learning to construct a class tightness graph, which serves as a regularization term to prevent overfitting. Finally, we employ Shannon entropy to quantify the overall uncertainty within the probability distribution. To validate the effectiveness of the proposed method, we conduct experiments using a multi-view epileptic EEG dataset and compare it with several other classification algorithms. The extensive experimental results demonstrate that the proposed MVRLR method achieves higher classification accuracy. By applying the proposed method, we can more accurately identify epileptic EEG signals, providing robust support for early diagnosis and personalized treatment of epilepsy.

Delivering the right amount of information to the right person is vital on the tactical battlefield. With the increasing use of mobile devices by the military, delivering relevant information instantaneously to the warfighter becomes possible. However, large quantities of data are being generated constantly while the human processing power and communication channels are limited. Therefore, data must be processed so it can be evaluated against operational needs. This data is collected in multiple modalities include images, videos and field reports with multi-sensor data. Providing automated processing of unstructured information promises to effectively connect information processing with operational decision making, dramatically reducing the time needed to identify relevant information for mission planning and execution. We describe a multi-view learning technique that augments the feature set used by a classifier in one modality with entity relationships discovered in other modalities. To accommodate the limited computation power of field devices, mostly handhelds, the multi-view learning algorithm is low complexity. It applies to multiple modalities, leveraging many-to-many correspondences among different modalities. Experiments on image and text are presented in the paper which show more than 20% improvement over categorizing text or images independently. The categorized information is matched to the mission and task needs. Finally, relevant information needs to be transmitted via limited bandwidth negotiated from limited resources.

Multi-view learning aims to effectively learn from data represented by multiple independent sets of attributes, where each set is taken as one view of the original data. In real-world application, each view should be acquired in unequal cost. Taking web-page classification for example, it is cheaper to get the words on itself (view one) than to get the words contained in anchor texts of inbound hyper-links (view two). However, almost all the existing multi-view learning does not consider the cost of acquiring the views or the cost of evaluating them. In this paper, we support that different views should adopt different representations and lead to different acquisition cost. Thus we develop a new view-dependent cost different from the existing both class-dependent cost and example-dependent cost. To this end, we generalize the framework of multi-view learning with the cost-sensitive technique and further propose a Cost-sensitive Multi-View Learning Machine named CMVLM for short. In implementation, we take into account and measure both the acquisition cost and the discriminant scatter of each view. Then through eliminating the useless views with a predefined threshold, we use the reserved views to train the final classifier. The experimental results on a broad range of data sets including the benchmark UCI, image, and bioinformatics data sets validate that the proposed algorithm can effectively reduce the total cost and have a competitive even better classification performance. The contributions of this paper are that: (1) first proposing a view-dependent cost; (2) establishing a cost-sensitive multi-view learning framework; (3) developing a wrapper technique that is universal to most multiple kernel based classifier.

Semi-Supervised Learning (SSL) aims to improve the performance of models trained with a small set of labeled data and a large collection of unlabeled data. Learning multi-view representations from different perspectives of data has proved to be very effectively for improving generalization performance. However, existing semi-supervised multi-view learning methods tend to ignore the specific difficulty of different unlabeled examples, such as the outliers and noise, leading to error-prone classification. To address this problem, this paper proposes Robust Transductive Support Vector Machine (RTSVM) that introduces the margin distribution into TSVM, which is robust to the outliers and noise. Specifically, the first-order (margin mean) and second-order statistics (margin variance) are regularized into TSVM, which try to achieve strong generalization performance. Then, we impose a global similarity constraint between distinct RTSVMs each trained from one view of the data. Moreover, our algorithm runs with fast convergence by using concave–convex procedure. Finally, we validate our proposed method on a variety of multi-view datasets, and the experimental results demonstrate that our proposed algorithm is effective. By exploring large amount of unlabeled examples and being robust to the outliers and noise among different views, the generalization performance of our method show the superiority to single-view learning and other semi-supervised multi-view learning methods.

In this paper, we employ multiple Single-hidden Layer Feedforward Neural Networks for multi-view action recognition. We propose an extension of the Extreme Learning Machine algorithm that is able to exploit multiple action representations and scatter information in the corresponding ELM spaces for the calculation of the networks’ parameters and the determination of optimized network combination weights. The proposed algorithm is evaluated by using two state-of-the-art action video representation approaches on five publicly available action recognition databases designed for different application scenarios. Experimental comparison of the proposed approach with three commonly used video representation combination approaches and relating classification schemes illustrates that ELM networks employing a supervised view combination scheme generally outperform those exploiting unsupervised combination approaches, as well as that the exploitation of scatter information in ELM-based neural network training enhances the network’s performance.

Classification of epileptic electroencephalograms can provide strong technical support for early warning and monitoring of epilepsy. However, Electroencephalogram (EEG) has characteristics of specificity, multi-feature, and susceptibility to noise interference. Therefore, this paper proposes a joint low-rank representation and transfer multi-view learning (LRR-TML) approach for EEG epilepsy classification. The LRR-TML approach inherits the advantages of low-rank representation, filters redundant features and noise information, and focuses on the main feature and significant feature of EEG. The LRR-TML approach introduces the ideas of transfer learning and multi-view learning on the framework of low-rank representation, mines the linear correlation between different domains in the subspace through low-rank representation, and effectively transfers the discriminative knowledge of the source domain to the target domain. According to the principles of complementarity and consistency of views, the LRR-TML approach comprehensively describes EEG by integrating various features, maximizing the consistency between different views and minimizing the divergence between views. Experimental results on the Bonn dataset show that the LRR-TML approach achieves the best classification results among all comparison approaches.

In this paper, we propose a sparse tensor regression model for multi-view feature selection. Apart from the most of existing methods, our model adopts a tensor structure to represent multi-view data, which aims to explore their underlying high-order correlations. Based on this tensor structure, our model can effectively select the meaningful feature set for each view. We also develop an iterative optimization algorithm to solve our model, together with analysis about the convergence and computational complexity. Experimental results on several popular multi-view data sets confirm the effectiveness of our model.

Multi-view feature selection is an important research direction in the field of multi-view learning. To remove the negative effect of irrelevant and redundant information and select the important features from the multi-view data, in this paper, we propose a novel hierarchical unsupervised multi-view feature selection method, which integrates matrix decomposition and hierarchical regularization as a joint model. Specifically, to exploit the consistency information of multiple views for projection matrix, we project the multiple views into a latent basis space based on a matrix decomposition model. The hierarchical regularization containing the $l_{2,1}$-norm, dependence and Frobenius-norm is imposed on projection matrix to, respectively, exploit the row-level, dependency-level and view-level feature selection. We also develop an efficient optimization algorithm to optimize our method. Extensive experimental results on six popular multi-view datasets show the effectiveness and superiority of our method by comparing with the state-of-the-art methods.

To understand text contents better, many research efforts have been made exploring detection and classification of the semantic relation between a concept pair. As described herein, we present our study of a semantic relation classification task as a graph-based multi-view learning task. Semantic relation can be naturally represented from two views: entity pair view and context view. Then we construct a weighted complete graph for each view and a bipartite graph to combine information of different views. An instance's label score is propagated on each intra-view graph and inter-view graph. The proposed algorithm is evaluated using the Concept Description Language for Natural Language (CDL) corpus and SemEval-2007 Task 04 dataset. The experimental results validate its effectiveness.