Search

Multi view clustering methods have become a research hotspot with the increasing diversity of data acquisition techniques. However, most clustering methods underestimate the impact of noise and complementary structural information of the data. Moreover, they often ignore the reverse guidance of clustering results on the optimization process of low rank tensors. To address these issues, this paper proposed a multi-view clustering method based on structured tensor learning (MCSTL). First, it further denoised the initial representation tensor to enhance its accuracy and robustness. At the same time, it complementarily learnt local structure, global structure, and high-order correlation across different views, which improved the consistency between the representation tensor and the intrinsic cluster structure of the original data. Then, it learnt a unified feature matrix from the affinity matrix of cross-view information fusion, and utilized the implicit clustering structure information within it to inversely guide the optimization process of the representation tensor. Lastly, it imposed an orthogonal constraint on the feature matrix, which provided soft label information of the data and allows for a direct clustering interpretation of the model. The experimental results show the MCSTL performs well in all six clustering evaluation metrics, with 27 out of 30 metrics reaching the optimal level, fully verifying the effectiveness and superiority of the MCSTL method. [ABSTRACT FROM AUTHOR]

Copyright of Radio Communications Technology is the property of 54th Research Institute of CETC and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

To address the timely acquisition of channel state information in reconfigurable intelligent surface (RIS)-assisted millimetre wave (mmWave) multiple-input multiple-output (MIMO) systems, a channel estimation scheme based on tensor decomposition was proposed. Firstly, a channel training mechanism with low pilot overhead was designed using a few passive reflection units and constructing a phase shift matrix. Then, a non-iterative channel estimation algorithm was derived using tensor canonical polyadic decomposition with Vandermonde structure constraints. Theoretical analysis indicated that the minimum pilot overhead of the proposed scheme only depended on the product of the subchannel path numbers of the reflection links and exhibited low computational complexity. Simulation results further verify the superiority of the proposed scheme compared to other methods.

Copyright of Acta Scientiarum Naturalium Universitatis Sunyatseni / Zhongshan Daxue Xuebao is the property of Sun-Yat-Sen University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

To address the problem of data heterogeneity and unified resource description in the centralized and dispersed ser-vice process of intelligent manufacturing in cloud computing environments, based on the analysis of the existing cloud manufacturing resource service packaging and publishing technology, this paper proposed a tensor-based cloud resource service encapsulation and publishing strategy by leveraging the technical characteristics of tensor theory in describing heterogeneous data. Firstly, it classified cloud manufacturing resources and providing encapsulation methods for different resources. Next, it defined tensor models for different types of cloud manufacturing resources based on tensor theory, established low-order sub-tensors according to different resource service properties, and used the tensor extension operator to fuse the low-order sub-tensors, embedded various attribute information into the high-order tensor space to achieve unified representation of cloud manufacturing resources. It provided a unified encapsulation template and formal description for manufacturing resources, manufacturing capabilities, and design resources. Finally, it proposed a cloud manufacturing resource service deployment and publishing strategy, designed and developed a cloud manufacturing resource service encapsulation and publishing system, and validated the effectiveness of the proposed method through examples. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The mechanical pressure mode of pressure expansion water seal brings the shortcomings ofcomplexity and non-durability of gate structure. In order to overcome these disadvantages, a kind of mountain type self-closing water seal which relied on the large deformation of it self to seal water was innovated. The key to the design of the mountain type water seal was to accurately predict the deformation of the water seal under various working conditions in the design stage. Based on the nonlinear tensor constitutive relation of water seal rubber material, the nonlinear characteristics of geometric large deformation, the contact nonlinear characteristics and the basic calculation principle of continuous medium, the simulation calculation of a mountain type self-closing water seal was carried out. The water sealing principle, water sealing law and effective water head to sealing water of mountain type self-closing water seal were discussed. The establishment of the type water seal can greatly reduce the construction cost of gate filling water seal device, facilitate engineering maintenance, improve durability and has great ap plication value. [ABSTRACT FROM AUTHOR]

Copyright of Journal of South China Normal University (Natural Science Edition) / Huanan Shifan Daxue Xuebao (Ziran Kexue Ban) is the property of Journal of South China Normal University (Natural Science Edition) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at the problem of the poor performance of the imputation method for the spreading codes blind estimation of the multi-antenna long-code direct sequence code division multiple access (CDMA) signals.Firstly,using the segmentation idea,the received signals were constructed into a three-dimensional tensor model,and the tensor was divided into sub-tensors according to the spreading gain.The sub-tensors were constructed into the Tucker decomposition model.Secondly,the spreading code fragments and the receiver gain matrixes were obtained from the sub-tensors by Tucker decomposition which was using the variable step gradient descending algorithm.Then,because of the same characteristic of the receiver gain matrixes in each sub-tensor,the permutation matrixes were estimated by using the cross correlation matrixes of the receiver gain matrixes and the order of the spreading code fragments was corrected.Finally,the amplitude of the spreading codes was determined by using the autocorrelation of the spreading codes,and each user’s spreading code was obtained.The simulation results verify the effectiveness of the proposed algorithm.

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Projection method was a traditional method for solving split feasibility problem. However,the solving efficiency of projection method depended on the selection of initial point and it was not directly applied to solve the tensor split feasibility problem. A new idea by using semidefinite relaxation algorithm to solve a kind of specific tensor split feasibility problem was proposed. The form of sets was given by polynomials inequalities,so the tensor split feasibility problem was relaxed to semidefinite program problem. The semidefinite relaxation method was used to solve the problem. The correlation theory of the semidefinite relaxation method was given and numerical experiments were carried out. The experimental results show that for different ranges of sets, different element values of tensors, different dimensions and different orders of tensors, the relaxation method can be used to solve the tensor split feasibility problem. [ABSTRACT FROM AUTHOR]

Copyright of Journal of South China University of Technology (Natural Science Edition) is the property of South China University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Chongqing University of Posts & Telecommunications (Natural Science Edition) is the property of Chongqing University of Posts & Telecommunications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Systems Engineering & Electronics is the property of Journal of Systems Engineering & Electronics Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Transactions of Beijing Institute of Technology is the property of Beijing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Jilin University (Science Edition) / Jilin Daxue Xuebao (Lixue Ban) is the property of Zhongguo Xue shu qi Kan (Guang Pan Ban) Dian zi Za zhi She and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal on Communication / Tongxin Xuebao is the property of Journal on Communications Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In order to solve the minimum number of multivariate polynomials and the global optimal value of the unconstrained polynomial optimization problem (POP), this paper first gives the proof of the conjecture of Liqun and Koklay that a 2n-degree or 2n+1-degree polynomial of r variables has at most nr isolated local minima when n≤2 . Secondly, since the problem of unconst-rained polynomial optimization is generally nonconvex, NP hard and its global optimal value is not easy to be solved, so by using the relevant knowledge of tensor, the theory and the proof of the lower bound of the global optimal value of the second order unconstrained multivariate polynomials are given. This theory is simple and convenient, which can be used to calculate the global optimal value better. [ABSTRACT FROM AUTHOR]

Tensor is a kind of data structure, which is basically multi-way array. Many data can he represented by tensor, such as face image. Feature extraction process is the most important part of face recognition while subsequent matching and identification process is based on it. TTrlSVD is a novel kind of tensor decomposition algorithm that can be thought as the extension of the matrix SVD towards tensor realm. The specific mode of the image in the image database is usually larger than the other modes, it is combined with TTrlSVD algorithm can further get the HOSVD to accelerate the facial feature extraction procedure. The organizations of the pictures were changed in the databases, which could be organized into different forms of tensor. TTrlSVD based tensor decomposition algorithm could extract facial features more efficiently and maintain good accuracy. Experimental results show that the proposed algorithm is more flexible and efficient than conventional tensor decomposition algorithms using Tensor Toolbox. [ABSTRACT FROM AUTHOR]

For the complexity of internal structure and the high dimension of influencing factors of stock market index, this paper proposed a method for time series forecasting of stock market index based on MPCA-RBF (multilinear principal component analysis-radial basis function) model. Due to the relationships between stock markets,this method selected seven stock markets and thirty-four stock technical indexes to construct a three-dimensional tensor model. Then, it utilized a tensor method MPCA for feature extraction of the model to preserve the internal structure of data while reducing dimension. Finally, by using RBF neural network to conduct regression prediction, the prediction accuracy was improved. Results on Hang Seng index and Nikkei 225 index closing price showed that,compared with non-tensor models, the forecast error of the proposed model was smaller and the forecasting accuracy was significantly improved. This shows that MPCA-RBF model can fully retain the internal structure of stock time series and it proves the validity and practicability of the model in field of stock forecasting. [ABSTRACT FROM AUTHOR]

With brain PET(positron emission tomography) image information， a recognition method based on hemisphere symmetry feature of tensor space was proposed to identify the epilepsy lesions of PET(positron emission tomography) images. Firstly， the SUV(standard uptake value) of each voxel in brain PET images was calculated and the third order tensor based on SUV was constructed. Then， the hemisphere symmetry feature was extracted and the hemisphere symmetry tensor model was built. Next， a multi linear principal component analysis (MPCA) algorithm was used for feature selection of hemisphere symmetry tensor model. Lastly， the support vector machine (SVM) was used to identify the epilepsy. The results show that the epilepsy lesions of the brain PET images can be effectively identified by the proposed algorithm， which can be used as a computer aided diagnosis way to help doctors with epilepsy disease diagnosis. [ABSTRACT FROM AUTHOR]

The method using the theory of low-rank matrix recovery! LHMR) based on convex optimization to model the background can't extract the moving object effectively when the background is unstable. Since matrix representation of video damaged the original temporal and spatial structure of data, this paper proposed an approach of moving object extraction based on iterative tensor high-order singular value decomposition (HOSVD), which utilized the tensor to characterize the high dimensional spatiotemporal structure of video. The proposed approach used HOSVD to replace the singular value decomposition (SVD) in LRMR, and utilized the augmented Lagrange multipliers to reconstruct the background portion and moving object portion of three-dimensional video tensor. Additionally, the proposed approach applied morphological opening and closing operations to the moving objects. Comparing with traditional methods, the experimental results show the proposed method can extract the moving objects more accurately and completely with lower misclassification rate. [ABSTRACT FROM AUTHOR]

Traditional methods of natural language processing input a good deal of handmade features into statistics learning models to process the text. At present, conditional random field model performed well in many natural language processing tasks, but the way of handmaking features and huge number of features made it more difficult to build the model, reduced the speed of model computing, and generated overfitting phenomenon. To solve these problems mentioned above, this paper introduced a novel conditional random field model extended by tensor. It designed complicated features automatically, reduced the workloads of handmaking features, used the Tucker decomposition algorithm to speed the model, and could be applied in many natural language processing tasks. Experiments show that the model can perform better in many natural language processing tasks than traditional conditional random field model if using the same elementary features. [ABSTRACT FROM AUTHOR]

The thermal buckling equation for thin spherical shells was deduced on the basis of axisymmetric thin spherical shell buckling equation derived with the tensor method. The thermal buckling equation involving the coupling of uniform external pressure and temperature was expressed in terms of displacement. The thin spherical shell buckling of minimum potential energy functional was also established according to the virtual work principle. 3 thermal buckling problems for simply supported hemispherical shells were analyzed with the Ritz method. The following 3 conclusions are drawn： 1) The critical value of uniform external pressure on condition that the temperature does not exceed the critical buckling level. 2) The buckling critical temperature value on condition that the uniform external pressure is 0. 3) The buckling critical temperature value on condition that the uniform external pressure does not exceed the critical level. [ABSTRACT FROM AUTHOR]

Traditional dimensionality reduction methods in face recognition are methods that reshape tensor face into a vector, which may lose the structural characteristics of the original data, leading to a relatively low identification result. We present a dimensionality reduction method------multilinear discriminant subspace projection (MDSP) based on tensor. Our algorithm aims to use tensor to describe face data directly, and project the tensor data onto the vector discriminant subspace through a new kind of projection method------tensor to vector projection (TVP). To reach this target, the algorithm first finds out the projection vectors (PV) that make data in the discriminant subspace get the maximum between-class scatter as well as the minimum within-class scatter. Then with the help of PV, tensor data can be projected into the low dimensional vector data. As long as proper constraints are given, the vector data can be the most representative feature data. The feature data is then sent to the KNN classifier for classification. Results in experiments on databases ORL confirm the veracity of our algorithm. [ABSTRACT FROM AUTHOR]

To establish the tensor model for color morphological operators, the paper proposed counter-harmonic mean-based color morphological operators by translating three-dimensional vectors into two-order positive definite symmetric tensors from HSV color space to tensor space. It used reference tensor method and tensorial similarity measure to construct input-preserving morphological operators. Finally, it conducted morphological gradient-based edge detection experiments to prove the feasibility and effectiveness of the proposed model. Experimental results show compared with traditional vector-based method, the tensor-based operators can get better effect. [ABSTRACT FROM AUTHOR]

To effectively restore the missing value in images, this paper proposed the iterative tensor high-order singular value decomposition (HOSVD) method. To simplify the solving process, the method divided the tensor nuclear norm objective function into sub-problems based on the Lagrange multiplier method. And it solved the sub-problems by using tensor HOSVD thresholding approach iteratively. Finally, it recovered the missing value. The proposed method can make full use of the multiple constrain relations among inter-frame and intra-frame images which can greatly improve the recovery precision. Simulation and real image experiment results demonstrate that this method has good restoration properties of missing value. [ABSTRACT FROM AUTHOR]

To solve the problem that tensile instability may occur in adopting Smoothed Particle Hydrodynamics (SPH) in the cases involved in material strength, an artificial stress of continuum type was presented to improve the tensile instability. The condition of application of the artificial stress was given and the tensor form of the stress was set up. The unification form of artificial stress was established through combining the artificial viscosity stress that eliminates compressive instability and the artificial stress that resists tensile instability. Two examples were calculated and the results were compared. It is indicated that better results can be obtained by using the artificial stress presented than those of other forms. The results have less vibration anti approach those of Finite Element Method, which proves that the artificial stress presented can improve the tensile instability in SPH. [ABSTRACT FROM AUTHOR]