Search

• We propose an improved differential evolution (DE) algorithm by composite binary crossover and Pareto refining operations to the DE. • DE solves the quantifying fraudulent transactions problem by minimizing the selected feature cost while maximizing the quantification ability. • We show that the proposed DE outperforms four other variants of DE and five state-of-the-art EAs on various credit datasets. • The results are also statically validated through the Friedman test and Wilcoxon post-hoc analysis. [Display omitted] Identification of fraudulent credit card transactions is a complex problem mainly due to the following factors: 1) The relative behavior of customers and fraudsters may alter over time. 2) The ratio of legitimate to fraudulent transactions is highly imbalanced, and 3) Investigators examine a small segment of transactions in a reasonable time frame. Researchers have proposed various algorithms to identify potential fraud in a new incoming transaction. However, these approaches require significant human investigator effort and are sometimes misleading. To address this issue, this paper proposes an improved multiobjective differential evolution (DE) algorithm to estimate the distribution of fraudulent transactions in a set of new incoming transactions, referred to as quantifying fraudulent transactions. Our paper has three major novelties. First, we present the problem formulation of cost-based feature selection with maximum quantification ability. Second, we improve the DE by applying effective trial vector generation algorithms to the random control parameter settings to exploit the advantage of individual DE variants. Third, we develop the maximum-relevancy-minimum-redundancy-based Pareto refining operator to enhance the self-learning ability of individuals in Pareto solutions. We compare our approach against four other modifications of DE and five state-of-the-art evolutionary algorithms on real-time credit datasets in streaming and non-streaming frameworks using hyper-volume, two-set coverage, and spread performance metrics. [ABSTRACT FROM AUTHOR]

Data selection is commonly used to reduce costs and energy usage by training on a subset of available data. However, determining the appropriate subset size requires extensive dataset knowledge and experimentation, limiting transferability. Varying the validation set also produces unstable results and wastes computational resources. In this paper, we propose a data selection method for dynamically determining subset ratios based on model performance using only a training set. The data search space is narrowed through weighted sampling, leveraging statistical selection patterns. Parallel analysis of class distributions identifies the most representative samples with high selection potential. Extensive experiments validate our approach and demonstrate improved training efficiency. Our method speeds up various subset ratios by up to 2.2x on CIFAR-10, 1.9x on CIFAR-100, 2.0x on TinyImageNet, and 2.3x on ImageNet with negligible accuracy drops. [Display omitted] Main flowchart of Dyna m ic Data S u bset S election for I mage C lassification (Music) to speed up neural network training, where reinforced data selection is performed every L epochs. Music is compatible with any data selection algorithm, and model parameters are updated using selected data subsets. The block diagram shows interactions between modules in the reinforced selection component. • Propose a data selection method to dynamically determine the subset ratio. • Study the role of intelligent data subset ratios. • Orthogonalize existing subset selection methods. • Two novel modules are used to narrow the data search. • Enable more efficient and sustainable AI development. [ABSTRACT FROM AUTHOR]

• The currently most comprehensive geometric features are established for facial beauty analysis. • The angle features as a new kind of geometric feature are proposed for facial beauty analysis. • The first optimized dense facial landmark model where each landmark has contribution to the facial beauty could achieve the current best facial beauty estimation performance with only geometric features. Existing geometric features on facial attractiveness analysis only focus on the ratios and distances, which is incomplete to represent all the information of a face. In this paper, we introduce a new category of feature, i.e., the angle features, to describe the angle of different organs such as the chin and eyes, which help boost the analysis performance in experiment. In addition, existing facial beauty analysis papers usually apply existing landmark models and extract their own different geometric feature sets on the landmarks. On the one hand, the geometric features are quite chaotic between different papers. On the other hand, most of the landmarks in the existing landmark model are useless for geometric feature extraction which wastes a lot of computational resources. To tackle these issues, we suggest to define a common geometric feature set and learn a special landmark model for attractiveness analysis. Specially, we collect all the available geometric features from the previous jobs and introduce a genetic feature selection algorithm to select the most effective geometric features. Furthermore, we introduce a special landmark model which exactly covers all the extracted geometric features. The experiments show that our method with the introduced angle features and the common feature set can outperform state-of-art facial beauty estimation methods with geometric features. [ABSTRACT FROM AUTHOR]

• We introduce a general framework to incorporate attribute selection into GNNs to enhance the compacity and robustness of GNNs. • We develop two kinds of GNNs, i.e., AsGCN and AsGAT for attributed graph data representation and learning. • We derive a simple and effective algorithm to optimize the proposed AsGCN and AsGAT. Graph Neural Networks (GNNs) have been widely used for graph learning tasks. The main aspect of GNN's layer-wise message passing is conducting attribute/feature propagation on graph. Most existing GNNs generally conduct feature propagation across all feature dimensions. However, in many real applications, attributes usually contain irrelevant and redundant noise. In this case, attribute/feature selection is desired to extract meaningful features and eliminate noisy ones for GNN's layer-wise propagation. Based on this observation, in this paper, we combine ℓ 2 , 1 / ℓ 1 -norm regularized attribute selection and GNNs together and propose a novel Attribute selection guided GNNs (AsGNNs) for graph data representation. AsGNNs aim to adaptively select some desired meaningful features/attributes that best serve GNNs. Moreover, an effective optimization framework has also been derived to train the proposed AsGNNs. The proposed AsGNNs provide a general framework which can incorporate any GNNs to conduct feature selection for layer-wise propagation. In this paper, we implement AsGNNs on both graph convolutional network (GCN) and graph attention network (GAT) and develop AsGCN and AsGAT for graph learning. Experimental results on several benchmark datasets demonstrate the effectiveness of the proposed AsGNNs (AsGCN, AsGAT) on semi-supervised learning tasks. [ABSTRACT FROM AUTHOR]

Multi-label feature selection plays an increasingly important role in alleviating the high dimensionality of multi-label learning tasks. Most extant methods posit that the learning task is performed in an environment where the label space is statically known. In reality, however, the environment is open and the labels may arrive dynamically, which is coined as streaming labels. Streaming labels-based multi-label feature selection suffers from many challenges derived from label space: (1) The label space expands dynamically; (2) Newly arrived labels exhibit complex relationships, often involving label correlation and labeling-importance. To cope with this challenge, in this paper, an intuitive yet effective algorithm named LLSL, i.e. learning implicit labeling-importance and label correlation for multi-label feature selection with streaming labels, is proposed. To be specific, the implicit labeling-importance with respect to streaming labels is firstly formalized by conducting the nearest neighbor reconstruction on feature space. Secondly, label correlation is seamlessly integrated into the objective function of feature relevance by designing the feature relevance influence factor. Based on the above, we build a feature conversion, which can realize the fusion of label-specific features for each streaming label. Finally, extensive experiments conducted on fifteen benchmark datasets provide clear evidence that LLSL has superior performance compared to three established streaming label-based MFS algorithms and seven static label space-based MFS algorithms. • By jointly learning the implicit labeling-importance for streaming labels and the impact of label correlation on feature importance, a novel framework is established to figure out multi-label feature selection with dynamic streaming labels. • The implicit labeling-importance for dynamic streaming labels is excavated by utilizing the neighborhood structure of samples on feature space, thereby providing valuable additional information for the learning task. • Label correlation is leveraged to advance generalization performance, especially, the effect degree of the already-arrived labels on the relationship between features and the newly-arrived labels is quantified to facilitate the model training. [ABSTRACT FROM AUTHOR]

• This paper proposes a method to replace the linear mapping-based spectral analysis with neural networks, which is more appropriate to learn the nonlinear mapping relationship between the original data and the pseudo cluster label. • Through constructing two different adaptive graph constraints, the mutual interaction between the NNs-based pseudo label learning module and the self-expression module is facilitated to explore the feature selection priority for the raw data. • Through a lot of well-designed experiments, the effectiveness of the method proposed in this paper has been fully proven. Unsupervised feature selection (UFS), which selects the most important feature subset and eliminates the unnecessary information for the upcoming data analysis, is a significant problem in machine learning and has been explored for years. Most UFS methods map features into a pseudo label space by multiplying a projection matrix constrained with sparsity to learn the mapping from the features to the labels. However, the mapping relationship is usually not linear, and linear regression may result in a suboptimal selection. To address this issue, we propose a novel UFS method, called neural networks embedded self-expression (NNSE). NNSE replaces the linear regression of traditional spectral analysis methods with neural networks to learn the pseudo label space. Besides, we embed neural networks into the self-expression model to improve the representative ability by preserving the local structure with an adaptive graph regularization module. Then we propose an efficient alternative iterative algorithm to solve the proposed model. Experimental results on 8 public datasets show NNSE outperforms the other state-of-the-art methods. Moreover, experimental results are also presented to show the convergence of the proposed method. The source code is available at: https://github.com/misteru/NNSE. [ABSTRACT FROM AUTHOR]

• The first attempt to design a GEP-based codematrix generation algorithm. The individual structure of GEP allows the codematrix to be optimized in a much faster and more effective way. • The feature selection scheme is combined with the design of codematrix, to accomplish the feature selection and codematrix optimization simultaneously. • A greedy strategy-based adaptive optimization method to further improve the individual's quality to deal with hard classes by appending proper columns to the codematrix. The multi-class classification task is one of the most common tasks in machine learning. As a typical solution based on a partitioning strategy, Error-Correcting Output Codes (ECOC) can transform a multi-class classification problem into multiple binary classification problems. The key of ECOC is to construct an effective codematrix to represent a set of class decomposition schemes, which transforms a multiclass problem into a group of binary class problems. Consequently, the design of a fast and effective ECOC codematrix generation method is of great research significance and value for solving multi-class classification problems. In ECOC algorithms, the design of codematrix is treated as a combination problem between different code columns, in which the evolutionary algorithm shows a great advantage. Based on this consideration, the Gene Expression Programming (GEP) is applied to search for the codematrix with high performance because its expressive tree structure makes it well represent codematrcies for subsequent optimization operations. This paper proposes an adaptive ECOC algorithm based on Gene Expression Programming (GEP) and similarity measurement matrix, named GEP-ECOC. In our GEP, each individual represents a set of columns to form a random ECOC codematrix, which is optimized in the evolutionary process. Meanwhile, the crossover and mutation operations are modified to include a legality checking process to ensure that the generated codematrix satisfies the ECOC constraints. The GEP-based ECOC codematrix generation algorithm can quickly produce a codematrix with better performance, which ensures the efficiency of the algorithm to a certain extent. In addition, an adaptive algorithm based on a similarity measurement matrix is proposed to add new columns to the current codematrix, aiming to better handle hard classes. Our algorithm is compared with other algorithms on various data sets, and the experimental results confirm that our GEP-ECOC can balance the efficiency and performance of the algorithm and achieve higher performance. [ABSTRACT FROM AUTHOR]

To utilize both labeled data and unlabeled data in real-world applications, semi-supervised learning is widely used as an effective technique. However, most semi-supervised methods do not perform well when there are many noises and redundant information in the original data. To address these issues, in this paper, we proposed a novel approach called robust embedding regression (RER) for semi-supervised learning by inheriting the advantages of the existing semi-supervised learning, robust linear regression, and low-rank representation techniques. Specifically, RER constructs a more robust and accurate graph by adaptively arranging the weight coefficient for each data point. Furthermore, the low-rank representation is introduced to reduce the negative influence of the redundant features and noises residing in the original data while the graph construction. More importantly, the proper norms are imposed on both the reconstruction and regularization terms to further improve the robustness and earn feature/sample selection. We designed an effective iterative algorithm to optimize the problem of RER. Comprehensive experimental results conducted on both synthetic and real-world datasets indicate that RER is superior in classification and clustering performance and robust to different types of noise compared with the existing semi-supervised methods. • The optimal norms in the proposed objective function have been chosen for robustness. • RER as a novel semi-supervised method has been strengthened functionally. • An optimization algorithm has been developed to solve the problem of RER. • Several experiments demonstrate the superiority of RER. [ABSTRACT FROM AUTHOR]

Landsat image segmentation is important for obtaining large-scale land cover maps. The accuracy of CNN-based Landsat image segmentation highly depends on the quantity and quality of the training samples. However, enough accurate labels for Landsat images are difficult to access. Fortunately, traditional classifier induced segmentation results can be considered as an alternative, although they are noisy and unbalanced to a certain extent. To resist noisy labels and alleviate the impact of imbalanced samples, this paper proposes a confidence interval based balanced random learning strategy. Firstly, a confidence interval-based mask is employed to control the random learning rate of the network from the entire noisy training set. Then, the multi-layer feature maps of CNN are fully utilized to compensate for the information loss in random learning, in which down-sampled labels are used to decrease the uncertainty brought by up-sampling CNN feature maps. In addition, considering the corruption of noisy labels on different classes, a balanced random learning with different confidence levels is performed on each class to further improve the learning ability of CNN. Experimental results on two widely used backbones, namely VGGNet and ResNet, demonstrate that the proposed balanced random learning strategy can effectively improve the performance of CNN under imbalanced and noisy labels, which can be improved by 3.41%. • Propose a random learning strategy to resist the impact of noisy labels. • Propose to compensate the information loss of random learning by multi-layer information. • Propose to use different confidence intervals on different classes to balance the learning ability. [ABSTRACT FROM AUTHOR]

• Using ℓ 20 norm constraint to obtain a pure sparse subspace. • Providing an algorithm to obtain a closed-form solution for ℓ 20 norm constraint problem. • The high efficiency of algorithm is guaranteed by the convergence proof. • Exploiting local discriminative structure to learn a better subspace. Clustering is a fashion method applied in machine learning tasks. However, high dimensional data brings many obstacles for clustering approaches. To address such a problem, the unsupervised feature selection (UFS) method can be incorporated into clustering to reduce dimensionality. In general, most of the UFS methods adopt ℓ 2 , 1 -norm for subspace sparsity learning. However, its sparsity highly relies on the setting of trade-off parameter, which may lead to instability of ranking results and the difficulty in obtaining the optimal solution of projection matrix. In this paper, we propose to directly learn an absolutely row-sparsity subspace via the ℓ 2 , 0 -norm constraint, called Sparse constraint and Local learning for Unsupervised Feature Selection (SLUFS). It is an ideal sparse subspace constraint which can overcome the drawbacks of the ℓ 2 , 1 -norm. However, optimizing the ℓ 2 , 0 -norm constraint is an NP-hard problem, and at present, only some approximate solutions can be given, but the convergence can not be guaranteed. To tackle this challenge, we design a novel alternative iterative algorithm to directly optimize the ℓ 2 , 0 -norm based model. Most importantly, our strategy can obtain a closed-form solution with strict convergence guarantee. Comprehensive experiments are conducted on several real-world datasets to evaluate the performance of SLUFS with comparison to several related state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• RUFSDR is proposed based on data relationship learning. It significantly improves the robustness in the scenario where outliers distributed both randomly and concentratedly are widely present. • A new metric that decides whether a sample is an outlier is introduced. It only depends on the numbers of nearest and farthest neighbors. Compared with the euclidean distance in high dimensional space, the numbers of neighbors are easier to understand. • Through data relationship learning, RUFSDR preserves a better intrinsic structure of the original data. Therefore, unsupervised feature selection and outlier removal can be conducted simultaneously. Unsupervised feature selection robust to many outliers is a challenging task. The crucial difficulty is learning a robust subspace, which preserves local structure. The most common solution is to reduce fitting error by applying different robust norms. However, there are three shortcomings. Firstly, they are not robust enough when outliers distributed both randomly and concentratedly are widely present. Secondly, outlier removal is not considered. Thirdly, it is not easy to understand and choose an euclidean distance threshold that decides a sample as an outlier in different scenarios. The first two shortcomings make previous methods fail to achieve their expected learning results, and the third one increases the application difficulty in different fields. To address these issues, a robust unsupervised feature selection via data relationship learning (RUFSDR) is proposed in this paper. Specifically, scores representing the data's importance will be learned and assigned to each sample. Inliers will be given different positive scores. Outliers will be given 0 such that a subspace, which preserves the local structure better, can be learned without prior knowledge about the distance threshold. The experiments conducted on various datasets with several scenarios show the superiority of RUFSDR. [ABSTRACT FROM AUTHOR]

· A novel method (RPI-CapsuleGAN) to predict RNA-protein interactions (RPI). · Use the elastic net feature selection method to screen high-correlation features as model input features to reduce the complexity of the model operation and avoid overfitting. · The convolutional attention mechanism is introduced into a generative adversarial capsule network to construct a classifier model for the first time. · Predict and analyze RPI network, and analyze the interpretability of RNA-protein interaction network. · RPI-CapsuleGAN increases the prediction performance on independent testing datasets compared with other state-of-the-art methods. RNA-protein interactions (RPI) play a crucial regulatory role in cellular physiological processes. The study and prediction of RPIs can be insightful for exploring disease mechanisms and drug target design. Traditional RPI prediction methods relied mainly on tedious and expensive biological experiments, and there is an increasing interest in developing more cost-effective computational methods to predict RPIs. This work proposes an interpretable RPI-CapsuleGAN method for RPI prediction based on a generative adversarial capsule network with a convolutional block attention module. First, RPI-CapsuleGAN extracts and fuses multiple features to characterize RNA and protein sequences. Subsequently, the elastic net feature selection method is used to retain features that are highly informative to RPI prediction. Finally, we introduce a convolutional attention mechanism into the generative adversarial capsule network for the first time in order to construct the RPI prediction framework, which is shown to improve the model feature learning of interpretable and expression ability, and effectively solves the problem of the disappearance of the model spatial structure hierarchy. Based on a five-fold cross-validation test, the prediction accuracy of the RPI-CapsuleGAN method reaches 97.1%, 88.8%, 92.5%, 97.3%, and 87.8% for datasets RPI488, RPI369, RPI2241, RPI1807, and RPI1446. The RPI-CapsuleGAN method has higher accuracy than state-of-the-art RPI prediction methods that use the same datasets. In the test dataset NPInter227 constructed in this paper, five groups of test sets are composed of positive samples and five groups of negative samples, the prediction accuracy reaches 97.38%, 96.48%, 97.38%, 97.81%, and 97.15%, respectively, outperforming other mainstream deep learning algorithms. In addition, RPI-CapsuleGAN obtained better results for the prediction of independent test datasets. Extensive experiments detailed here show that RPI-CapsuleGAN can provide an efficient, accurate, and stable method for RPI prediction. [ABSTRACT FROM AUTHOR]

• We propose a multilabel feature selection method that considers the trade-off between the relevance of the features and the feature cost. • We derive a score function specialized as a cost-constrained multilabel feature selection problem based on normalization. • We identify the bound of the cost factor between the relevance of the features and the importance of the feature cost. • We devise a new framework to handle the shortage of datasets containing feature costs. Feature selection is one of the key steps in building a predictive model in multi-label classification. However, most of the existing methods do not take into account information about the costs associated with considered features, such as the costs of performing diagnostic medical tests. We consider a problem of cost-constrained multilabel feature selection, which aims to select a feature subset relevant to multiple labels while satisfying a user-specific maximal admissible budget. This approach allows for building a model with high predictive power, for which the cost of making a prediction for a single instance does not exceed the user-specified budget. In this problem, the balance between the feature subset relevance and its cost should be considered concurrently, which is nontrivial in practice because their optimal balance is unknown. In this paper, we propose a novel criterion for cost-constrained multilabel feature selection that combines the relevance and cost of the candidate feature. The relevance measure is derived using the lower bound of the mutual information between the feature subset and label vector. Moreover, we propose an effective method for determining the cost-factor value that controls the trade-off between relevancy and cost. The experimental results on multilabel datasets with various characteristics demonstrate the superiority of the proposed method over conventional methods. [ABSTRACT FROM AUTHOR]

• Novel holistic transformer vector-representation network for modeling scenes/agents. • Simultaneous output multi-modal trajectory prediction and behavior/lane decision. • Organically integrated system of attention mechanisms based on multiple cues. • Attention-weighted regularized trajectory prediction using behavior decision tasks. Trajectory prediction and behavioral decision-making are two important tasks for autonomous vehicles that require a good understanding of the environmental context. Notably, behavioral decisions are better made by referring to the outputs of trajectory predictions. However, most current solutions perform these tasks separately. Therefore, this paper proposes a new joint holistic transformer network that combines multiple cues to predict trajectories and make behavioral decisions simultaneously. To better explore the intrinsic relationships among cues, the network uses existing knowledge and adopts three kinds of attention mechanisms: the sparse multi-head type for reducing noise impact, feature selection sparse type for optimally using partial prior knowledge, and multi-head with sigmoid activation type for optimally using posteriori knowledge. Compared with other trajectory prediction models, the proposed model has a better comprehensive performance and good interpretability. Perceptual noise robustness experiments demonstrate that the proposed model has good noise robustness. Thus, simultaneous trajectory prediction and behavioral decision-making combining multiple cues are accomplished, which reduces computational costs and enhances semantic relationships between scenes and agents. [ABSTRACT FROM AUTHOR]

COVID-19, as an infectious disease, has shocked the world and still threatens the lives of billions of people. Early detection of COVID-19 patients is an important issue for treating and controlling the disease from spreading. In this paper, a new strategy for detecting COVID-19 infected patients will be introduced, which is called Distance Biased Naïve Bayes (DBNB). The novelty of DBNB as a proposed classification strategy is concentrated in two contributions. The first is a new feature selection technique called Advanced Particle Swarm Optimization (APSO) which elects the most informative and significant features for diagnosing COVID-19 patients. APSO is a hybrid method based on both filter and wrapper methods to provide accurate and significant features for the next classification phase. The considered features are extracted from Laboratory findings for different cases of people, some of whom are COVID-19 infected while some are not. APSO consists of two sequential feature selection stages, namely; Initial Selection Stage (IS2) and Final Selection Stage (FS2). IS2 uses filter technique to quickly select the most important features for diagnosing COVID-19 patients while removing the redundant and ineffective ones. This behavior minimizes the computational cost in FS2, which is the next stage of APSO. FS2 uses Binary Particle Swarm Optimization (BPSO) as a wrapper method for accurate feature selection. The second contribution of this paper is a new classification model, which combines evidence from statistical and distance based classification models. The proposed classification technique avoids the problems of the traditional NB and consists of two modules; Weighted Naïve Bayes Module (WNBM) and Distance Reinforcement Module (DRM). The proposed DBNB tries to accurately detect infected patients with the minimum time penalty based on the most effective features selected by APSO. DBNB has been compared with recent COVID-19 diagnose strategies. Experimental results have shown that DBNB outperforms recent COVID-19 diagnose strategies as it introduce the maximum accuracy with the minimum time penalty. [ABSTRACT FROM AUTHOR]

• A new feature selection approach is proposed, which combines and utilizes multiple individual methods in order to achieve a more generalized solution. • A hybrid solution is proposed in the paper which combines the given, available (supervised, state-of-the-art) feature selection techniques that have their own specific, but fixed feature evaluation measures/metrics. • Adaptivity of the proposed algorithm is realized in such a way that at an individual step of the feature selection algorithm it iterates not only in the space of the variables but in the space of available features selection techniques, too. • Different directions of tests were applied: linear and non-linear dependencies with varying data distribution, noise and outliers; using benchmark datasets of the UCI Machine Learning Repository and also own real-life datasets; comparison to recent state-of-the-art feature selection methods. • T The proposed AHFS nearly doubles the accuracy (resulting in around half value for the related modeling error) compared to the individual methods, making it a superior feature selection algorithm . This paper deals with the problem of integrating the most suitable feature selection methods for a given problem in order to achieve the best feature order. A new, adaptive and hybrid feature selection approach is proposed, which combines and utilizes multiple individual methods in order to achieve a more generalized solution. Various state-of-the-art feature selection methods are presented in detail with examples of their applications and an exhaustive evaluation is conducted to measure and compare the their performance with the proposed approach. Results prove that while the individual feature selection methods may perform with high variety on the test cases, the combined algorithm steadily provides noticeably better solution. [ABSTRACT FROM AUTHOR]

• Adaptive anchor neighbor graph is learned to attenuate the influence of the mismatch between the neighbor structure and the category relationship in the high-dimensional data manifold structure. • A sparse constraint via ℓ 2 , 0 -norm is adopted to guarantee the sparsity of spectral analysis model;, and the orthogonal constraint of projection matrix guarantees subspace orthogonality to improve model performance. • An equivalent form of the original model is proposed, an iterative optimization algorithm is presented to deal with the non-convex optimization problem in fewer iterations and less time. Feature selection has been a research hotspot in many fields. Models based on graph learning are currently the most popular approaches. However, the sparsity of most models is not strong, and graph learning for pair-sample evaluation takes a lot of time. ℓ 2 , 1 -norm regularization is the sparsity strategy adopted in most sparse models at present since the convex function is easy to solve. Nevertheless, the sparsity of ℓ 2 , 1 -norm is insufficient, and there exist parameter adjustment problems. ℓ 2 , 0 -norm is a better choice, which can strengthen the sparse constraints of the subspace. In this paper, the Sparse feature selection via Fast Embedding Spectral Analysis (SFESA) is proposed. Firstly, an adaptive anchor nearest neighbor graph is constructed to avoid the high time cost of learning pairwise nearest neighbor graphs to a certain extent. The low-dimensional embedding of data manifold structure is maintained by performing spectral analysis for the constructed graph. Secondly, the projected data is approximated to the low-dimensional embedding structure via a regularization term. Finally, ℓ 2 , 0 -norm is employed to constrain the projection matrix to enhance the subspace sparsity. Furthermore, a fast iterative algorithm is presented to solve this non-convex optimization problem. Extensive experiments on multiple public datasets show that SFESA can obtain excellent performance in less time. [ABSTRACT FROM AUTHOR]

• We introduce a density-based clustering algorithm applying reverse nearest-neighbours. • Unlike other density algorithms, ours calculates cluster-specific feature weights. • We show its superior cluster recovery on data sets with and without noise features. DBSCAN is arguably the most popular density-based clustering algorithm, and it is capable of recovering non-spherical clusters. One of its main weaknesses is that it treats all features equally. In this paper, we propose a density-based clustering algorithm capable of calculating feature weights representing the degree of relevance of each feature, which takes the density structure of the data into account. First, we improve DBSCAN and introduce a new algorithm called DBSCANR. DBSCANR reduces the number of parameters of DBSCAN to one. Then, a new step is introduced to the clustering process of DBSCANR to iteratively update feature weights based on the current partition of data. The feature weights produced by the weighted version of the new clustering algorithm, W-DBSCANR, measure the relevance of variables in a clustering and can be used in feature selection in data mining applications where large and complex real-world data are often involved. Experimental results on both artificial and real-world data have shown that the new algorithms outperformed various DBSCAN type algorithms in recovering clusters in data. [ABSTRACT FROM AUTHOR]

• The supervised similarity measure based on information theory incorporating the information of class labels is applied to quantify similarities between features. • The proposed feature selection problem via holistic redundancy minimization based on the supervised similarity measure is mathematically formulated as a biconvex optimization problem with a quartic objective function. In addition, an iteratively reweighted convex quadratic program is reformulated. • A two-timescale duplex neurodynamic system is applied to solve the formulated biconvex optimization problem and a projection neural network is customized to solve the iteratively reweighted convex optimization problem. • The two-timescale duplex neurodynamic approach is substantiated to be almost-surely convergent to global optimal solutions to the formulated biconvex optimization problem. • Extensive experimental results on benchmark datasets demonstrate the superior performance of the proposed QWRM-based feature selection methods in comparison with the mainstream feature selection methods. Feature selection is an important dimensionality reduction technique in machine learning, pattern recognition, image processing, and data mining. Most existing feature selection methods are greedy in nature thus are prone to sub-optimality. Though some feature selection methods based on global optimization of unsupervised redundancy may potentiate performance improvements, they may or may not be relevant to classification as the information on pairwise features with class labels is missing. In this paper, based on a supervised similarity measure, a biconvex optimization problem is formulated for holistic feature section with a quadratically weighted objective function subject to linear equality and nonnegativity constraints. In addition, an iteratively reweighted convex quadratic program is reformulated. A two-timescale duplex neurodynamic system is applied to solve the formulated biconvex optimization problem and a projection neural network is customized to solve the iteratively reweighted convex optimization problem. Experimental results of the proposed neurodynamics-based supervised feature selection are elaborated in comparison with several existing feature selection methods based on twenty benchmark datasets to substantiate the efficacy and superiority of the neurodynamics-based method for selecting informative features in classification. [ABSTRACT FROM AUTHOR]

• We propose a new PBDR method to generate an ideal block diagonal representation matrix for large-scale data, which is significantly quicker than traditional block diagonal-based subspace clustering methods. • To illustrate the effectiveness of our proposed method, we theoretically and empirically validate the block diagonal effect and feasibility guarantee. • Extensive experiments have been conducted on several benchmark datasets, and it illustrates that our method can efficiently handle large-scale data without sacrificing clustering accuracy. Block Diagonal Representation (BDR) has attracted massive attention in subspace clustering, yet the high computational cost limits its widespread application. To address this issue, we propose a novel approach called Projective Block Diagonal Representation (PBDR), which rapidly pursues a representation matrix with the block diagonal structure. Firstly, an effective sampling strategy is utilized to select a small subset of the original large-scale data. Then, we learn a projection mapping to match the block diagonal representation matrix on the selected subset. After training, we employ the learned projection mapping to quickly generate the representation matrix with an ideal block diagonal structure for the original large-scale data. Additionally, we further extend the proposed PBDR model (i. e. , PBDR ℓ 1 and PBDR *) by capturing the global or local structure of the data to enhance block diagonal coding capability. This paper also proves the effectiveness of the proposed model theoretically. Especially, this is the first work to directly learn a representation matrix with a block diagonal structure to handle the large-scale subspace clustering problem. Finally, experimental results on publicly available datasets show that our approaches achieve faster and more accurate clustering results compared to the state-of-the-art block diagonal-based subspace clustering approaches, which demonstrates its practical usefulness. [ABSTRACT FROM AUTHOR]

• Our point-based refinement module only considers the uncertain points rather than all the points when performing matching on the finer scales. • Our adaptive matching module which flexibly assigns each memory feature point with an appropriate matching scheme. • The proposed module can better adapt to video contexts and achieve better complementary between different matching schemes. • Our PMNet achieves the state-of-the-art performance on several benchmark datasets while retaining competitive efficiency. Recent years have witnessed the prevalence of memory-based methods for Semi-supervised Video Object Segmentation (SVOS) which utilise past frames efficiently for label propagation. When conducting feature matching, fine-grained multi-scale feature matching has typically been performed using all query points, which inevitably results in redundant computations and thus makes the fusion of multi-scale results ineffective. In this paper, we develop a new Point-based Memory Network, termed as PMNet, to perform fine-grained feature matching on hard samples only, assuming that easy samples can already obtain satisfactory matching results without the need for complicated multi-scale feature matching. Our approach first generates an uncertainty map from the initial decoding outputs. Next, the fine-grained features at uncertain locations are sampled to match the memory features on the same scale. Finally, the matching results are further decoded to provide a refined output. The point-based scheme works with the coarsest feature matching in a complementary and efficient manner. Furthermore, we propose an approach to adaptively perform global or regional matching based on the motion history of memory points, making our method more robust against ambiguous backgrounds. Experimental results on several benchmark datasets demonstrate the superiority of our proposed method over state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Streaming feature selection (SFS), is the task of selecting the most informative features in dealing with high-dimensional or incrementally growing problems. Several SFS algorithms have been proposed in the literature. However, they do not consider all feature subsets at the redundancy analysis step due to computational concerns. Moreover, they do not reconsider previously removed features which leads to losing most of the useful information. In this paper, the redundancy analysis step is defined as a binary optimization problem. Then, a binary bat algorithm (BBA) is adopted to find the minimal informative subsets. In this way, a large number of feature subsets can be considered effectively at the redundancy analysis step. In addition, an effective priority list is used to maintain previously removed redundant features. Such a list allows the re-examination of informative features. As a result, it is possible to consider the mutual information between features that are not streamed in an small time interval. Experimental studies on fifteen different types of datasets show that our approach is superior to state-of-the-art online and offline streaming feature selection methods in terms of classification accuracy. [ABSTRACT FROM AUTHOR]

• A unified model for sparse optimal scoring is proposed by employing L u -norm (0 ≤ q ≤ 1) regular term where L o − norm and(L u) -norm (0 ≤ q ≤ 1) will be selected adaptively to find more sparser solutions. • We derive an efficient iterative algorithm based on alternating direction method of multipliers (ADMM) for the new formulation. The new proposed method can solve (L o −) norm regularized problem directly rather than using convex or nonconvex approximations of (L o −) norm. • The convergence of the algorithm is analyzed theoretically. • Extensive numerical experiments show that our algorithm is efficient not only in classification accuracy but also in sparsity. Optimal scoring (OS), an equivalent form of linear discriminant analysis (LDA), is an important supervised learning method and dimensionality reduction tool. However, it is still a challenge for the classical OS on small sample size (SSS) datasets. In this paper, to find sparse discriminant vectors, we propose a unified model for sparse optimal scoring (SOS) by virtue of the generalized ℓ q -norm (0 ≤ q ≤ 1). To overcome the difficulty in treating the generalized ℓ q -norm, we propose an efficient alternative direction method of multipliers (ADMM), where proximity operator of ℓ q -norm is employed for different q values. Meanwhile, the convergence results of our method are also established. Numerical experiments on artificial and benchmark datasets demonstrate the effectiveness and feasibility of our proposed method. [ABSTRACT FROM AUTHOR]

• SLREO performs feature selection in the latent representation space, uses latent representation learning to mine the hidden information between data, and retains the interconnection between data. • SLREO generates pseudo-label information through the OLSDA method embedded in a non-negative manifold structure. • The l 2,1 -norm constraint is imposed on the residual matrix of latent representation learning to ensure the robustness of the clustering indicators. • Unifying the latent representation matrix and the clustering index matrix can preserve the dependencies between data and ensure non-negative of pseudo-labels. • Applying uncorrelated constraint and l 2,1 -norm constraint on the feature transformation matrix can avoid excessive suppression of non-zero rows and the appearance of redundant solutions. And more discriminative features can be selected. Modern unsupervised feature selection methods predominantly obtain the cluster structure and pseudo-labels information through spectral clustering. However, the pseudo-labels obtained by spectral clustering are usually mixed between positive and negative. Moreover, the Laplacian matrix in spectral clustering typically affects feature selection. Additionally, spectral clustering does not consider the interconnection information between data. To address these problems, this paper proposes uncorrelated feature selection via sparse latent representation and extended orthogonal least square discriminant analysis (OLSDA), which we term SLREO). Firstly, SLREO retains the interconnection between data by latent representation learning, and preserves the internal information between the data. In order to remove redundant interconnection information, an l 2,1 -norm constraint is applied to the residual matrix of potential representation learning. Secondly, SLREO obtains non-negative pseudo-labels through orthogonal least square discriminant analysis (OLSDA) of embedded non-negative manifold structure. It not only avoids the appearance of negative pseudo-labels, but also eliminates the effect of the Laplacian matrix on feature selection. The manifold information of the data is also preserved. Furthermore, the matrix of the learned latent representation and OLSDA is used as pseudo-labels information. It not only ensures that the generated pseudo-labels are non-negative, but also makes the pseudo-labels closer to the true class labels. Finally, in order to avoid trivial solutions, an uncorrelated constraint and l 2,1 -norm constraint are imposed on the feature transformation matrix. These constraints ensure row sparsity of the feature transformation matrix, select low-redundant and discriminative features, and improve the effect of feature selection. Experimental results show that the Clustering Accuracy (ACC) and Normalized Mutual Information (NMI) of SLREO are significantly improved, as compared with six other published algorithms, tested on 11 benchmark datasets. [ABSTRACT FROM AUTHOR]

• A hybrid Machine Learning (ML) approach for an efficient wrapper feature selection is proposed for enhancing the performance and reducing the computation time of IDS while increasing the accuracy to detect the intruder in IoT networks. • The hybrid ML method consists of two techniques: Differential Evolution (DE) and Extreme Learning Machine (ELM). The DE is responsible for choosing the beneficial features while the ELM is employed to assess the selected features. • Extensive experiments are implemented to evaluate the proposed approach using different performance metrics. In addition, to verify the efficiency of proposed model, the comparison is achieved with some related existing methods such as CFS + SVM [20] , CFS + ANN [19] , GA + Bagging [18] , CFS + RF [30] , and Chi-squared + RF [30]. The intrusion detection system (IDS) has gained a rapid increase of interest due to its widely recognized potential in various security fields, however, it suffers from several challenges. Different network datasets have several redundant and irrelevant features that affect the decision of the IDS classifier. Therefore, it is essential to decrease these features to improve the system performance. In this paper, an efficient wrapper feature selection method is proposed for improving the performance and decreasing the processing time of the IDS. The proposed approach employs a differential evaluation algorithm to select the useful features whilst the extreme learning machine classifier is applied after feature selection to evaluate the selected features. Many experiments are performed using the full NSL-KDD dataset to evaluate the performance of the proposed method. The results prove that the proposed approach can efficiently reduce the features, increase the accuracy, reduce the false alarm rates, and improve the processing time of the IDS in comparison to other recent related works. [ABSTRACT FROM AUTHOR]

• A novel texture descriptor to address both scale transformation and noise interference. • An extended LBP with lightweight feature dimension. • Maintains both macro and micro descriptive information in the spatial and spectral domains. • Outperforms thirty classical LPB variants as well as eight typical deep learning methods. • Experiments on five public databases and one fresh texture database. As one of the most successful local feature descriptors, the local binary pattern (LBP) estimates the texture distribution rule of an image based on the signs of differences between neighboring pixels to obtain intensity- and rotation- invariance. In this paper, we propose a novel image descriptor to address scale transformation and noise interference simultaneously. We name it scale-selective and noise-robust extended LBP (SNELBP). First, each image in training sets is transformed into different scale spaces by a Gaussian filter. Second, noise-robust pattern histograms are obtained from each scale space by using our previously proposed median robust extended LBP (MRELBP). Then, scale-invariant histograms are determined by selecting the maximum among all scale levels for a certain image. Finally, the most informative patterns are selected from the dictionary pretrained by the two-stage compact dominant feature selection method (CDFS), maintaining the descriptor more lightweight with sufficiently low time cost. Extensive experiments on five public databases (Outex_TC_00011, TC_00012, KTH-TIPS, UMD and NEU) and one fresh texture database (JoJo) under two kinds of interferences (Gaussian and salt pepper) indicate that our SNELBP yields more competitive results than thirty classical LPB variants as well as eight typical deep learning methods. [ABSTRACT FROM AUTHOR]

• Mutual information feature selection method based on conditional mutual information. • Novel mutual information feature selection algorithm, exploring high order dependencies. • The proposed method allows seeing the mutual information feature selection with highorder dependencies with clear interpretation. • Compared with different methods over 20 benchmark datasets, the proposed method reached the best results. • The proposed method is faster and accurate than other feature selection methods that explore high order dependencies. This paper presents a novel feature selection method based on the conditional mutual information (CMI). The proposed High Order Conditional Mutual Information Maximization (HOCMIM) method incorporates high order dependencies into the feature selection procedure and has a straightforward interpretation due to its bottom-up derivation. The HOCMIM is derived from the CMI's chain expansion and expressed as a maximization optimization problem. The maximization problem is solved using a greedy search procedure, which speeds up the entire feature selection process. The experiments are run on a set of benchmark datasets (20 in total). The HOCMIM is compared with eighteen state-of-the-art feature selection algorithms, from the results of two supervised learning classifiers (Support Vector Machine and K-Nearest Neighbor). The HOCMIM achieves the best results in terms of accuracy and shows to be faster than high order feature selection counterparts. [ABSTRACT FROM AUTHOR]

• A new strategy for predicting the behavior of the person's body if he has been infected with Covid-19, which is called Covid-19 Prudential Expectation Strategy (CPES) is introduced. • CPES has the ability to classify people based on their bodies' reaction to Covid-19 infection, which is built upon a proposed Statistical Naïve Bayes (SNB) classifier. • For enhancing the classification accuracy, CPES employs two proposed techniques for outlier rejection and feature selection, which are called Hybrid Outlier Rejection (HOR) method and Improved Binary Genetic Algorithm (IBGA) method respectively. • HOR rejects outliers in the training data using a hybrid method that combines standard division and Binary Gray Wolf Optimization (BGWO) methods. • On the other hand, IBGA selects the most useful features for the prediction process using hybrid method that consists of Fisher Score (F Score) as a fast method and BGA as an accurate method that depends on the average accuracy value from several classification models as a fitness function. • CPES has been compared against recent related technologies for Covid-19 diagnosing. • CPES outperforms the competing methods in experimental results because it provides the best results with values of 0.87, 0.13, 0.84, and 0.79 for accuracy, error, precision, and recall. Covid-19, what a strange, unpredictable mutated virus. It has baffled many scientists, as no firm rule has yet been reached to predict the effect that the virus can inflict on people if they are infected with it. Recently, many researches have been introduced for diagnosing Covid-19; however, none of them pay attention to predict the effect of the virus on the person's body if the infection occurs but before the infection really takes place. Predicting the extent to which people will be affected if they are infected with the virus allows for some drastic precautions to be taken for those who will suffer from serious complications, while allowing some freedom for those who expect not to be affected badly. This paper introduces Covid-19 Prudential Expectation Strategy (CPES) as a new strategy for predicting the behavior of the person's body if he has been infected with Covid-19. The CPES composes of three phases called Outlier Rejection Phase (ORP), Feature Selection Phase (FSP), and Classification Phase (CP). For enhancing the classification accuracy in CP, CPES employs two proposed techniques for outlier rejection in ORP and feature selection in FSP, which are called Hybrid Outlier Rejection (HOR) method and Improved Binary Genetic Algorithm (IBGA) method respectively. In ORP, HOR rejects outliers in the training data using a hybrid method that combines standard division and Binary Gray Wolf Optimization (BGWO) method. On the other hand, in FSP, IBGA as a hybrid method selects the most useful features for the prediction process. IBGA includes Fisher Score (F Score) as a filter method to quickly select the features and BGA as a wrapper method to accurately select the features based on the average accuracy value from several classification models as a fitness function to guarantee the efficiency of the selected subset of features with any classifier. In CP, CPES has the ability to classify people based on their bodies' reaction to Covid-19 infection, which is built upon a proposed Statistical Naïve Bayes (SNB) classifier after performing the previous two phases. CPES has been compared against recent related strategies in terms of accuracy, error, recall, precision, and run-time using Covid-19 dataset [1]. This dataset contains routine blood tests collected from people before and after their infection with covid-19 through a Web-based form created by us. CPES outperforms the competing methods in experimental results because it provides the best results with values of 0.87, 0.13, 0.84, and 0.79 for accuracy, error, precision, and recall. [ABSTRACT FROM AUTHOR]

• A number of information-theoretic based feature selection approaches (ITFSs) are reviewed in terms of feature correlation. For comparison purposes, the correlation they focus on and their objective evaluation functions are listed. • This paper is the first work to comprehensively investigate and define multiple correlations, including relevance, redundancy, complementarity, and interaction in the feature selection process within the framework of information theory. The distinctions and connections between multiple correlations are explored in depth at four different levels. • A new feature selection approach that combines multiple correlations of features is constructed and corresponding feature selection algorithm with class-based relevance, redundancy, com-plementarity, and interaction (R2CI) is designed. • The results of comparisons and hypothesis testing against eleven related feature selection algorithms on twenty datasets show that the proposed algorithm has significant advantages in most cases. Information theoretic-guided feature selection approaches (ITFSs), which exploit the uncertainty of information to measure the correlation of features, aim to select the most informative features. However, most previous approaches suffer from two drawbacks. 1) Complementarity and interaction are not valued, leading to features with potential discriminatory information for learning tasks such as classification not being excavated and affecting the effectiveness of learning. 2) The various correlations that exist between features for the class have not been fully considered, and their differentiation and relationships have not been well reflected. To address the former issue, guided by information theory, the complementarity and interaction between features are studied. For the latter, firstly, some ITFSs are reviewed and analyzed in terms of feature correlation. The analysis reveals that considering feature multi-correlation is absent in the selection process. Motivated by this problem, a feature selection algorithm with class-based relevance, redundancy, complementarity, and interaction (R2CI) is designed for the first time. Moreover, the distinctions and connections among different correlations are also explored. The results of comparisons and hypothesis test against competitive algorithms show that R2CI has significant advantages in most cases. [ABSTRACT FROM AUTHOR]

• An adaptive-similarity-based multi-modality feature selection method is proposed. • This method captures the intrinsic similarity shared across different modality data. • This method discovers the disease sensitive biomarkers. • This method yields better performance on identifying disease status. Multimodal classification methods using different modalities have great advantages over traditional single-modality-based ones for the diagnosis of Alzheimer's disease (AD) and its prodromal stage mild cognitive impairment (MCI). With the increasing amount of high-dimensional heterogeneous data to be processed, multi-modality feature selection has become a crucial research direction for AD classification. However, traditional methods usually depict the data structure using pre-defined similarity matrix as a priori, which is difficult to precisely measure the intrinsic relationship across different modalities in high-dimensional space. In this paper, we propose a novel multimodal feature selection method called Adaptive-Similarity-based Multi-modality Feature Selection (ASMFS) which performs adaptive similarity learning and feature selection simultaneously. Specifically, a similarity matrix is learned by jointly considering different modalities and at the same time, an efficient feature selection is conducted by imposing group sparsity-inducing l 2 , 1 -norm constraint. Evaluated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database with baseline MRI and FDG-PET imaging data collected from 51 AD, 43 MCI converters (MCI-C), 56 MCI non-converters (MCI-NC) and 52 normal controls (NC), we demonstrate the effectiveness and superiority of our proposed method against other state-of-the-art approaches for multi-modality classification of AD/MCI. [ABSTRACT FROM AUTHOR]

• Focusing on the directional outliers (DOs). • A Feature Selection via Directional Outliers Correcting (FSDOC) method for supervised feature selection is proposed. • An optimization algorithm captures directional outliers; two correcting algorithms capture redundant features. • Theorems for rationality and justification of algorithms. • The FSDOC method outperforms eight comparative methods on fifteen datasets. With the rapid development of multimedia technologies (e.g. deep learning), Feature Selection (FS) is now playing a critical role in acquiring discriminative features from massive data. Traditional FS methods score feature importance and select the top best features by treating all instances equally; Hence, valuable instances like directional outliers (DOs), which are specific outliers closer to other class centres than to their owns, seldom receive particular attention during feature selection. Based on our observation, DOs derive from "misclassified instances" which lead to misclassification. In this paper, we present a novel supervised feature selection method entitled Feature Selection via Directional Outliers Correcting (FSDOC), for accurate data classification. The proposed FSDOC includes an optimization algorithm to capture DOs, and two correcting algorithms to reasonably capture redundant features by correcting DOs with intraclass deviation minimization and interclass relative distance maximization. We give theoretical guarantees and adequate analysis on all algorithms to show the effectiveness of FSDOC. Extensive experiments on fifteen public datasets, and two case studies of deep features and very-high dimensional Fisher Vector selection, demonstrate the superior performance of FSDOC. [ABSTRACT FROM AUTHOR]

• We propose a novel two-stage aware training framework for siamese networks, in which position-aware and appearance-aware training schemes are presented to optimize the shallow and the deep network layers, respectively. This contribution helps siamese tracker to achieve precise and robust visual tracking. • An effective feature selection module is presented to solve the online adaptation problem of Siamese tracker. By analyzing the changing principle of feature distribution, the module combines diverse attention networks in a unique way to explore the real discriminative features for the current object. • The proposed tracker is evaluated on four popular benchmark datasets extensively. The results demonstrate that the tracker performs better than other state-of-the-art methods in terms of accuracy and robustness. Siamese networks have achieved great success in visual tracking with the advantages of speed and accuracy. However, how to track an object precisely and robustly still remains challenging. One reason is that multiple types of features are required to achieve good precision and robustness, which are unattainable by a single training phase. Moreover, Siamese networks usually struggle with online adaption problem. In this paper, we present a novel two-stage aware attentional Siamese network for tracking (Ta-ASiam). Concretely, we first propose a position-aware and an appearance-aware training strategy to optimize different layers of Siamese network. By introducing diverse training patterns, two types of required features can be captured simultaneously. Then, following the rule of feature distribution, an effective feature selection module is constructed by combining both channel and spatial attention networks to adapt to rapid appearance changes of the object. Extensive experiments on various latest benchmarks have well demonstrated the effectiveness of our method, which significantly outperforms state-of-the-art trackers. [ABSTRACT FROM AUTHOR]

• We proposed a novel LIS_StLSR method for robust image classification. • LIS_StLSR consists of a low-rank inter-class sparsity constraint. • A semi-flexible regression target is also proposed to measure the regression error. • LIS_StLSR obtains low-dimensional and highly-discriminant feature representation. • Proposed method achieves state-of-the-art performance in extensive experiments. Least squares regression (LSR) is an important machine learning method for feature extraction, feature selection, and image classification. For the training samples, there are correlations among samples from the same class. Therefore, many LSR-based methods utilize this property to pursue discriminative representation. However, if the training samples contain noise or outliers, it will be hard to obtain the exact inter-class correlation. To address this problem, in this paper, a novel LSR-based method is proposed, named low-rank inter-class sparsity based semi-flexible target least squares regression (LIS_StLSR). Firstly, the low-rank representation method is utilized to achieve the intrinsic characteristics of the training samples. Afterwards, the low-rank inter-class sparsity constraint is used to force the projected data to have an exact common sparsity structure in each class, which will be robust to noise and outliers in the training samples. This step can also reduce margins of samples from the same class and enlarge margins of samples from different classes to make the projection matrix discriminative. The low-rank representation and the discriminative projection matrix are jointly learned such that they can be boosted mutually. Moreover, a semi-flexible regression target matrix is introduced to measure the regression error more accurately, thus the regression performance can be enhanced to improve the classification accuracy. Experiments are implemented on the different databases of Yale B, AR, LFW, CASIA NIR-VIS, 15-Scene SPF, COIL-20, and Caltech 101, illustrating that the proposed LIS_StLSR outperforms many state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• We propose a novel RGB-D salient object detection model, which takes the equalities of input into consideration. • An SG-MWMG sub-network is designed to determine the informative and non-informative regions in the input RGB-D images. • An MCFF module is designed to fuse the unimodal RGB and depth features at multiple scales and levels. Most existing RGB-D salient object detectors make use of the complementary information of RGB-D images to overcome the challenging scenarios, e.g., low contrast, clutter backgrounds. However, these models generally neglect the fact that one of the input images may be poor in quality. This will adversely affect the discriminative ability of cross-modal features when the two channels are fused directly. To address this issue, a novel end-to-end RGB-D salient object detection model is proposed in this paper. At the core of our model is a Semantic-Guided Modality-Weight Map Generation (SG-MWMG) sub-network, producing modality-weight maps to indicate which regions on both modalities are high-quality regions, given input RGB-D images and the guidance of their semantic information. Based on it, a Bi-directional Multi-scale Cross-modal Feature Fusion (Bi-MCFF) module is presented, where the interactions of the features across different modalities and scales are exploited by using a novel bi-directional structure for better capturing cross-scale and cross-modal complementary information. The experimental results on several benchmark datasets verify the effectiveness and superiority of the proposed method over some state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• A novel unsupervised feature selection method is proposed which aims at preserving the local invariance of feature subspace by local linear embedding and manifold regularization simultaneously. • An alternating algorithm based on alternative direction method of multiplier s with computational complexity analysis is designed to handle the resulting optimization problem. • A variety of experiments on several publicly available real world datasets are conducted to demonstrate the effectiveness and superiority of the proposed method. As one of the important dimensionality reduction techniques, unsupervised feature selection (UFS) has enjoyed amounts of popularity over the last few decades, which can not only improve learning performance, but also enhance interpretability and reduce computational costs. The existing UFS methods often model the data in the original feature space, which cannot fully exploit the discriminative information. In this paper, to address this issue, we investigate how to strengthen the relationship between UFS and the feature subspace, so as to select relevant features more straightforwardly and effectively. Methodologically, a novel UFS approach, referred to as Graph Regularized Local Linear Embedding (GLLE), is proposed by integrating local linear embedding (LLE) and manifold regularization constrained in feature subspace into a unified framework. To be more specific, we explicitly define a feature selection matrix composed of 0 and 1, which can realize the process of UFS. For the purpose of modelling the feature selection matrix, we propose to preserve the local linear reconstruction relationship among neighboring data points in the feature subspace, which corresponds to LLE constrained in the feature subspace. To make the feature selection matrix more accurate, we propose to use manifold regularization as an assistant of LLE to find the relevant and representative features such that the selected features can make each sample under the feature subspace be accordance with the manifold assumption. A tailored iterative algorithm based on Alternative Direction Method of Multipliers (ADMM) is designed to solve the proposed optimization problem. Extensive experiments on twelve real-world benchmark datasets are conducted, and the more promising results are achieved compared with the state-of-the-arts approaches. [ABSTRACT FROM AUTHOR]

• The area under the analyst certainty curve is a better measure compared to the minimum feature prefix. • For the outlier-based sequential explanations (SE), the particle swarm optimisation search method only outperforms the greedy search methods that make use of the same outlier scoring measure. • The sample-based SE, support vector machine recursive feature elimination SE, returned the best performing SE overall. • The best performing outlier and sample-based SEs outperformed the best performing sequential feature explanations (SFE). In most applications, anomaly detection operates in an unsupervised mode by looking for outliers hoping that they are anomalies. Unfortunately, most anomaly detectors do not come with explanations about which features make a detected outlier point anomalous. Therefore, it requires human analysts to manually browse through each detected outlier point's feature space to obtain the subset of features that will help them determine whether they are genuinely anomalous or not. This paper introduces sequential explanation (SE) methods that sequentially explain to the analyst which features make the detected outlier anomalous. We present two methods for computing SEs called the outlier and sample-based SE that will work alongside any anomaly detector. The outlier-based SE methods use an anomaly detector's outlier scoring measure guided by a search algorithm to compute the SEs. Meanwhile, the sample-based SE methods employ sampling to turn the problem into a classical feature selection problem. In our experiments, we compare the performances of the different outlier- and sample-based SEs. Our results show that both the outlier and sample-based methods compute SEs that perform well and outperform sequential feature explanations. [ABSTRACT FROM AUTHOR]

Many embedded feature selection methods ignore the correlation among the important features. To reduce correlation, some models introduce constraints to impose sparsity on features, some try to exploit the similarity and group features without changing the objective function. In this paper, we propose diverse feature selection (DFS), which simultaneously performs feature clustering and selection. Given a dataset with known class labels, we separate the features into a set of feature clusters where the features in the same cluster have a higher correlation with each other than with the features in different clusters. A diverse regularization (DR) is proposed to reduce the linear and nonlinear correlations among important features. Using this regularization, DFS can select features that are both informative and diverse. The experimental results on seven image datasets, five gene datasets as well as four other datasets demonstrate the superior performance of DFS. [ABSTRACT FROM AUTHOR]

• Presenting a new multi-label feature selection method that efficiently combines manifold regularization and dependence maximization. • Introducing a HSIC-based measurement to evaluate the dependence between the manifold space and label space. • Developing an iterative optimization method to solve the objective function of our method MRDM with good convergence. • Conducting extensive experiments on various multi-label data sets to demonstrate the superiority of the proposed method. Feature selection is able to select more discriminative features for classification and plays an important role in multi-label learning to alleviate the effect of the curse of dimensionality. Recently, the multi-label feature selection methods based on the sparse regression model have received increasing attentions. However, most of these methods directly project original data space to label space in the regression model, which is inappropriate because the linear assumption between data space and label space doesn't hold in most cases. In the paper, we propose a feature selection method named multi-label feature selection via manifold regularization and dependence maximization (MRDM). In the regression model of MRDM, the original data space is projected to a low-dimensional manifold space, which not only has the same topological structure with the original data, but also has a strong dependence with the class labels. Then, an objective function involving l 2 , 1 -norm regularization is formulated, and an alternating optimization-based iterative algorithm is designed to obtain the sparse coefficients for multi-label feature selection. Extensive experiments on various multi-label data sets demonstrate the superiority of the proposed method compared with some state-of-the-art multi-label feature selection methods. [ABSTRACT FROM AUTHOR]

• Usage of population and neighbourhood based metaheuristic search techniques for feature selection, • Application of fuzzy rough set for evaluation of subsets of features as the objective function, and • Showcasing the effectiveness of hybridized global and local feature selection. Cancer is a cluster of diseases caused due to unusual cell growth. This paper aims to discover cancer prediction from the microarray gene expression data using the selected features. The metaheuristic search algorithms select the global and local optimal features using population and neighbourhood based algorithms. Although the ant colony optimization and genetic algorithm search for the global optimal features from the dataset entails enhanced classification, sometimes there occur some challenges in the selection of neighbourhood features. Against this background, two feature selection algorithms are proposed to hybridize tabu search, a neighbourhood based search algorithm with global optimal feature selection algorithm. Those are (1) Ant Colony Optimization and Tabu search with Fuzzy Rough set for Optimal feature selection (ACTFRO) algorithm, (2) Genetic algorithm and Tabu search with Fuzzy Rough set for Optimal feature selection (GATFRO) algorithm. The performance of proposed feature selection algorithms is assessed through a fuzzy rough nearest neighbour classifier using ten-fold cross validation. Four cancer medical datasets and one non-medical dataset are used to analyse the performance of the proposed algorithms in terms of classification accuracy, computation time, sensitivity, specificity, f-measure, receiver operation characteristics and positive predicted value. Results derived from the different performance metrics confirm that the proposed algorithms evidence effective global and local feature selection hybridization with improved results. [ABSTRACT FROM AUTHOR]

• We propose a new feature selection method based on graph-based feature representations and the fused lasso framework. • Our approach can accommodate structural relationship between pairs of samples through graph-based features. • Our method can enhance the trade-off between the relevance of each feature and the redundancy between pairwise features. • An iterative algorithm is developed to identify the most discriminative features. • Experiments demonstrate that our proposed approach can outperform its competitors on benchmark datasets. Most state-of-the-art feature selection methods tend to overlook the structural relationship between a pair of samples associated with each feature dimension, which may encapsulate useful information for refining the performance of feature selection. Moreover, they usually consider candidate feature relevancy equivalent to selected feature relevancy, and therefore, some less relevant features may be misinterpreted as salient features. To overcome these issues, we propose a new feature selection method based on graph-based feature representations and the Fused Lasso framework in this paper. Unlike state-of-the-art feature selection approaches, our method has two main advantages. First, it can accommodate structural relationship between a pair of samples through a graph-based feature representation. Second, our method can enhance the trade-off between the relevancy of each individual feature on the one hand and its redundancy between pairwise features on the other. This is achieved through the use of a Fused Lasso framework applied to features reordered on the basis of their relevance with respect to the target feature. To effectively solve the optimization problem, an iterative algorithm is developed to identify the most discriminative features. Experiments demonstrate that our proposed approach can outperform its competitors on benchmark datasets. [ABSTRACT FROM AUTHOR]

• Deep subspace clustering using self-representative network is proposed. • Feature extraction and emphasis network using attention model is introduced. • Dimensional reduction in self-expressive layer is presented. • We experimentally verify performance improvements for the proposed approach. Deep learning based subspace clustering networks have been a significant technique for motion segmentation, unsupervised image segmentation, image representation and compression, and face clustering by separating the high-dimensional data points into their representative low-dimensional linear subspaces. Effective feature selection is critical to remove redundant samples and select the representative feature subset from high-dimensional data space; hence deriving the number of subspaces, their dimensions, data segmentation, and a basis for each subspace. The effective self-representative feature selection and emphasis by scaling the feature map in the learned embedded space is required for deep learning based subspace clustering to reduce the number of parameters and dimension of the self-representative layer. In this paper, we propose a self-representative feature extraction deep neural network for unsupervised subspace clustering to improve representativeness and clustering ability. The extensive relevant results on various data demonstrate that deep subspace clustering employing self-representative features from high-dimensional data can effectively reduce the dimension of the self-representative layer while improving performance. [ABSTRACT FROM AUTHOR]

• We propose a new nonlinear feature selection method based on kernelized multiclass support vector machine (COMSVM). • The recursive feature elimination algorithm is improved by adding the batch elimination and the rescreening process. • The fast recursive feature elimination algorithm is implemented with the proposed COMSVM. • The experiment results demonstrate the efficiency of our new method. Feature selection is an important procedure in machine learning because it can reduce the complexity of the final learning model and simplify the interpretation. In this paper, we propose a novel non-linear feature selection method that targets multi-class classification problems in the framework of support vector machines. The proposed method is achieved using a kernelized multi-class support vector machine with a fast version of recursive feature elimination. The proposed method selects features that work well for all classes, as the involved classifier simultaneously constructs multiple decision functions that separates each class from the others. We formulate the classifier as a large optimisation problem, and iteratively solve one decision function at a time, leading to a lower computational time complexity than when solving the large optimisation problem directly. The coefficients of the classifier are then used as a ranking criterion in the accelerated recursive feature elimination by adding batch elimination and a rechecking process. Experimental results on several datasets demonstrate the superior performance of the proposed feature selection method. [ABSTRACT FROM AUTHOR]

• A two-stage algorithm for high-dimensional feature selection is proposed. • A advanced hybrid ant colony optimization algorithm is proposed. • Our method has good selection performance with short running time. Ant colony optimization (ACO) is widely used in feature selection owing to its excellent global/local search capabilities and flexible graph representation. However, the current ACO-based feature selection methods are mainly applied to low-dimensional datasets. For thousands of dimensional datasets, the search for the optimal feature subset (OFS) becomes extremely difficult due to the exponential increase of the search space. In this paper, we propose a two-stage hybrid ACO for high-dimensional feature selection (TSHFS-ACO). As an additional stage, it uses the interval strategy to determine the size of OFS for the following OFS search. Compared to the traditional one-stage methods that determine the size of OFS and search for OFS simultaneously, the stage of checking the performance of partial feature number endpoints in advance helps to reduce the complexity of the algorithm and alleviate the algorithm from getting into a local optimum. Moreover, the advanced ACO algorithm embeds the hybrid model, which uses the features' inherent relevance attributes and the classification performance to guide OFS search. The test results on eleven high-dimensional public datasets show that TSHFS-ACO is suitable for high-dimensional feature selection. The obtained OFS has state-of-the-art performance on most datasets. And compared with other ACO-based feature selection methods, TSHFS-ACO has a shorter running time. [ABSTRACT FROM AUTHOR]