In this paper, we propose a new unsupervised spectral feature selection model by embedding a graph regularizer into the framework of joint sparse regression for preserving the local structures of ...data. To do this, we first extract the bases of training data by previous dictionary learning methods and, then, map original data into the basis space to generate their new representations, by proposing a novel joint graph sparse coding (JGSC) model. In JGSC, we first formulate its objective function by simultaneously taking subspace learning and joint sparse regression into account, then, design a new optimization solution to solve the resulting objective function, and further prove the convergence of the proposed solution. Furthermore, we extend JGSC to a robust JGSC (RJGSC) via replacing the least square loss function with a robust loss function, for achieving the same goals and also avoiding the impact of outliers. Finally, experimental results on real data sets showed that both JGSC and RJGSC outperformed the state-of-the-art algorithms in terms of k -nearest neighbor classification performance.
Display omitted
•Ni/Mg/Al layered double hydroxides (NMA-LDHs) synthesized.•NMA-LDHs with hierarchically hollow microsphere structure.•Calcined NMA-LDHs have large adsorption capacities for CR and ...Cr(VI) ions.
The preparation of hierarchical porous materials as catalysts and sorbents has attracted much attention in the field of environmental pollution control. Herein, Ni/Mg/Al layered double hydroxides (NMA-LDHs) hierarchical flower-like hollow microspheres were synthesized by a hydrothermal method. After the NMA-LDHs was calcined at 600°C, NMA-LDHs transformed into Ni/Mg/Al layered double oxides (NMA-LDOs), which maintained the hierarchical flower-like hollow structure. The crystal phase, morphology, and microstructure of the as-prepared samples were characterized by X-ray diffraction, scanning electron microscopy, transmission electron microscopy, energy-dispersive X-ray spectroscopy elemental mapping, Fourier transform infrared spectroscopy, and nitrogen adsorption−desorption methods. Both the calcined and non-calcined NMA-LDHs were examined for their performance to remove Congo red (CR) and hexavalent chromium (Cr(VI)) ions in aqueous solution. The maximum monolayer adsorption capacities of CR and Cr(VI) ions over the NMA-LDOs sample were 1250 and 103.4mg/g at 30°C, respectively. Thermodynamic studies indicated that the adsorption process was endothermic in nature. In addition, the addition of coexisting anions negatively influenced the adsorption capacity of Cr(VI) ions, in the following order: CO32−>SO42−>H2PO4−>Cl−. This work will provide new insight into the design and fabrication of advanced adsorption materials for water pollutant removal.
This paper proposes a new unsupervised spectral feature selection method to preserve both the local and global structure of the features as well as the samples. Specifically, our method uses the ...self-expressiveness of the features to represent each feature by other features for preserving the local structure of features, and a low-rank constraint on the weight matrix to preserve the global structure among samples as well as features. Our method also proposes to learn the graph matrix measuring the similarity of samples for preserving the local structure among samples. Furthermore, we propose a new optimization algorithm to the resulting objective function, which iteratively updates the graph matrix and the intrinsic space so that collaboratively improving each of them. Experimental analysis on 12 benchmark datasets showed that the proposed method outperformed the state-of-the-art feature selection methods in terms of classification performance.
Hashing is becoming increasingly important in large-scale image retrieval for fast approximate similarity search and efficient data storage. Many popular hashing methods aim to preserve the kNN graph ...of high dimensional data points in the low dimensional manifold space, which is, however, difficult to achieve when the number of samples is big. In this paper, we propose an effective and efficient hashing approach by sparsely embedding a sample in the training sample space and encoding the sparse embedding vector over a learned dictionary. To this end, we partition the sample space into clusters via a linear spectral clustering method, and then represent each sample as a sparse vector of normalized probabilities that it falls into its several closest clusters. This actually embeds each sample sparsely in the sample space. The sparse embedding vector is employed as the feature of each sample for hashing. We then propose a least variance encoding model, which learns a dictionary to encode the sparse embedding feature, and consequently binarize the coding coefficients as the hash codes. The dictionary and the binarization threshold are jointly optimized in our model. Experimental results on benchmark data sets demonstrated the effectiveness of the proposed approach in comparison with state-of-the-art methods.
•As far as we know, we are the first to propose a general framework to incorporate the quantization-based methods into the conventional similarity-preserving hashing, in order to improve the ...effectiveness of hashing methods. In theory, any quantization method can be adopted to reduce the quantization error of any similarity-preserving hashing methods to improve their performance.•This framework can be applied to both unsupervised and supervised hashing. We experimentally obtained the best performance compared to state-ofthe-art supervised and unsupervised hashing methods on six popular datasets.•We successfully show it to work on a huge dataset SIFT1B (1 billion data points) by utilizing the graph approximation and out-of-sample extension.
Nowadays, due to the exponential growth of user generated images and videos, there is an increasing interest in learning-based hashing methods. In computer vision, the hash functions are learned in such a way that the hash codes can preserve essential properties of the original space (or label information). Then the Hamming distance of the hash codes can approximate the data similarity. On the other hand, vector quantization methods quantize the data into different clusters based on the criteria of minimal quantization error, and then perform the search using look-up tables. While hashing methods using Hamming distance can achieve faster search speed, their accuracy is often outperformed by quantization methods with the same code length, due to the low quantization error and more flexible distance lookups. To improve the effectiveness of the hashing methods, in this work, we propose Quantization-based Hashing (QBH), a general framework which incorporates the advantages of quantization error reduction methods into conventional property preserving hashing methods. The learned hash codes simultaneously preserve the properties in the original space and reduce the quantization error, and thus can achieve better performance. Furthermore, the hash functions and a quantizer can be jointly learned and iteratively updated in a unified framework, which can be readily used to generate hash codes or quantize new data points. Importantly, QBH is a generic framework that can be integrated to different property preserving hashing methods and quantization strategies, and we apply QBH to both unsupervised and supervised hashing models as showcases in this paper. Experimental results on three large-scale unlabeled datasets (i.e., SIFT1M, GIST1M, and SIFT1B), three labeled datastes (i.e., ESPGAME, IAPRTC and MIRFLICKR) and one video dataset (UQ_VIDEO) demonstrate the superior performance of our QBH over existing unsupervised and supervised hashing methods.
In image analysis, the images are often represented by multiple visual features (also known as multiview features), that aim to better interpret them for achieving remarkable performance of the ...learning. Since the processes of feature extraction on each view are separated, the multiple visual features of images may include overlap, noise, and redundancy. Thus, learning with all the derived views of the data could decrease the effectiveness. To address this, this paper simultaneously conducts a hierarchical feature selection and a multiview multilabel (MVML) learning for multiview image classification, via embedding a proposed a new block-row regularizer into the MVML framework. The block-row regularizer concatenating a Frobenius norm (F-norm) regularizer and an l 2,1 -norm regularizer is designed to conduct a hierarchical feature selection, in which the F-norm regularizer is used to conduct a high-level feature selection for selecting the informative views (i.e., discarding the uninformative views) and the 12,1-norm regularizer is then used to conduct a low-level feature selection on the informative views. The rationale of the use of a block-row regularizer is to avoid the issue of the over-fitting (via the block-row regularizer), to remove redundant views and to preserve the natural group structures of data (via the F-norm regularizer), and to remove noisy features (the 12,1-norm regularizer), respectively. We further devise a computationally efficient algorithm to optimize the derived objective function and also theoretically prove the convergence of the proposed optimization method. Finally, the results on real image datasets show that the proposed method outperforms two baseline algorithms and three state-of-the-art algorithms in terms of classification performance.
•The Pt-TiO2 catalyst is deactivated by adsorption of halogen ions.•The halogen poison is mainly attributed to the active site blocking of the Pt surface.•Halogen ions and Pt form PtX coordination ...bonds.•Large halogen diameter exhibits severe poisoning effect.
Catalytic decomposition of formaldehyde (HCHO) at room temperature is an important method for HCHO removal. Pt-based catalysts are the optimal catalyst for HCHO decomposition at room temperature. However, the stability of this catalyst remains unexplored. In this study, Pt-TiO2 (Pt-P25) catalysts with and without adsorbed halogen ions (including F−, Cl−, Br−, and I−) were prepared through impregnation and ion modification. Pt-TiO2 samples with adsorbed halogen ions exhibited reduced catalytic activity for formaldehyde decomposition at room temperature compared with the Pt-TiO2 sample; the catalytic activity followed the order of F-Pt-P25, Cl-Pt-P25, Br-Pt-P25, and I-Pt-P25. Characterization results (including XRD, TEM, HRTEM, BET, XPS, and metal dispersion) showed that the adsorbed halogen ions can poison Pt nanoparticles (NPs), thereby reducing the HCHO oxidation activity of Pt-TiO2. The poison mechanism is due to the strong adsorption of halogen ions on the surface of Pt NPs. The adsorbed ions form coordination bonds with surface Pt atoms by transferring surplus electrons into the unoccupied 5d orbit of the Pt atom, thereby inhibiting oxygen adsorption and activation of the Pt NP surface. Moreover, deactivation rate increases with increasing diameter of halogen ions. This study provides new insights into the fabrication of high-performance Pt-based catalysts for indoor air purification.
A flux-reversal permanent magnet (FRPM) machine typically suffers from unfavorably large cogging torque due to its special doubly salient structure and high air-gap flux density, resulting in ...undesired torque and speed ripples, as well as acoustic noise and vibration, especially at low speeds. Therefore, an improved configuration of FRPM machines by introducing a small space-gap between the two adjacent magnets belonging to the same stator tooth is proposed in this paper. The analytical expression of cogging torque taking this small space gap into consideration is derived, and furthermore, the optimal dimensions of the space gap are obtained analytically for cogging torque minimization. In addition, the influences of the space-gap-based configuration on key electromagnetic performances, including cogging torque, back electromotive force, and electromagnetic torque, are investigated by both 2-D finite-element analysis and experimental results. It turns out to be that by employing this configuration, not only the cogging torque, and hence, torque ripples can be suppressed, but also the electromagnetic torque can be improved.
This paper proposes a new hashing framework to conduct similarity search via the following steps: first, employing linear clustering methods to obtain a set of representative data points and a set of ...landmarks of the big dataset; second, using the landmarks to generate a probability representation for each data point. The proposed probability representation method is further proved to preserve the neighborhood of each data point. Third, PCA is integrated with manifold learning to lean the hash functions using the probability representations of all representative data points. As a consequence, the proposed hashing method achieves efficient similarity search (with linear time complexity) and effective hashing performance and high generalization ability (simultaneously preserving two kinds of complementary similarity structures, i.e., local structures via manifold learning and global structures via PCA). Experimental results on four public datasets clearly demonstrate the advantages of our proposed method in terms of similarity search, compared to the state-of-the-art hashing methods.
Head injury is reported to be associated with increased risks of dementia and Alzheimer's disease (AD) in many but not all the epidemiological studies. We conducted a systematic review and ...meta-analysis to estimate the relative effect of head injury on dementia and AD risks.
Relevant cohort and case-control studies published between Jan 1, 1990, and Mar 31, 2015 were searched in PubMed, Web of Science, Scopus, and ScienceDirect. We used the random-effect model in this meta-analysis to take into account heterogeneity among studies.
Data from 32 studies, representing 2,013,197 individuals, 13,866 dementia events and 8,166 AD events, were included in the analysis. Overall, the pooled relative risk (RR) estimates showed that head injury significantly increased the risks of any dementia (RR = 1.63, 95% CI 1.34-1.99) and AD (RR = 1.51, 95% CI 1.26-1.80), with no evidence of publication bias. However, when considering the status of unconsciousness, head injury with loss of consciousness did not show significant association with dementia (RR = 0.92, 95% CI 0.67-1.27) and AD (RR = 1.49, 95% CI 0.91-2.43). Additionally, this positive association did not reach statistical significance in female participants.
The findings from this meta-analysis indicate that head injury is associated with increased risks of dementia and AD.