Yue Fang

China Academy of Railway Sciences

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Cong Lei

Tianjin University of Finance and Economics

Le Wang

Tianjin University

Shichao Zhang

Guangxi Normal University

Shengchun Wang

China Academy of Railway Sciences

Yangding Li

Changsha Normal University

Yonggang Li

Hubei Provincial Center for Disease Control and Prevention

Shengwei Ren

Henan Provincial Eye Hospital

Rongyao Hu

Hefei University of Technology

Shaoli Liu

Beijing Institute of Technology

Qiang Han

Civil Aviation University of China

Cooperative Institutions

China Academy of Railway Sciences

Beijing Institute of Technology

Guangxi Normal University

Beijing Jiaotong University

Chinese Academy of Sciences

Zhejiang University

Xi'an Jiaotong University

Shanghai Jiao Tong University

Ministry of Education of the People's Republic of China

University of Electronic Science and Technology of China

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Unsupervised Spectral Feature Selection with Local Structure Learning

Shichao Zhang Yue Fang Cong Lei Yangding Li Rongyao Hu

Because there are many unlabeld high-dimensional data need to be processed, traditional unsupervised feature selection becomes an important and challenging issue in machine learning domain. The graph matrix that makes the selected features rely on high levels of the learned structure is usually constructed by traditionary embedded unsupervised methods. Nevertheless, data from real life usually contain a lot of noise samples and redundant features that make the graph matrix produced by original data be untenable. This paper designs a novel feature selection algorithm to make the graph matrix credible for classifying high-dimensional data. The local structure learning and feature selection are simultaneously employed to determine the graph matrix adaptively. We make graph matrix contain more available data structure information by constraining the graph matrix, thus we use the graph matrix learning to maintain the local structure between the samples, and the proposed method can select significant features. Moreover, we derive an efficient and iterative update method to optimize the problem. The extensive empirical results demonstrate that the proposed method can select the most discriminative features and achieve the best classification performance, compared to the competing methods.

Discriminative model

Feature Learning

10.1109/icbk.2017.23

Cite

Citations (1)

Supervised Feature Selection Algorithm Based on Low-Rank and Manifold Learning

Lecture notes in computer science (2017)

Yue Fang Jilian Zhang Shichao Zhang Cong Lei Xiaoyi Hu

Rank (graph theory)

Manifold alignment

Manifold (fluid mechanics)

Feature (linguistics)

Intrinsic dimension

Representation

Feature Learning

10.1007/978-3-319-69179-4_19

Cite

Citations (3)

The Influence of Recovery Experience on Occupational Well-Being: A Moderated Mediating Effect

Frontiers in Business Economics and Management (2023)

Yue Fang HE Wen-hua

Based on the perspective of intergenerational differences, this study takes the questionnaire data of 660 enterprise employees as a sample to explore the boundary conditions of the path of “recovery experience → psychological capital → occupational well-being”. The results show that both transactional psychological capital and interpersonal psychological capital have significant mediating effects, and intergenerational differences do play a negative moderating role in the path of “recovery experience → psychological capital → occupational well-being”. And further research shows that the indirect effect of interpersonal psychological capital of Generation Z employees is not significant. Overall, this study complements the research on the boundary conditions of the path of “recovery experience → psychological capital → occupational well-being”, and provides a reference for the effective management of the new generation of employees in practice.

Positive psychological capital

Capital (architecture)

Sample (material)

10.54097/fbem.v9i2.9000

Cite

Citations (1)

Construction of Cloud ERP Security Evaluation Index System Based on Text Mining

Frontiers in artificial intelligence and applications (2020)

Ying Gao Qin Su Yue Fang Pinyan Lai

The aim of this study is to investigate and discuss the potential security issues arising from the deployment of the Cloud Enterprise Resource Planning system, and to propose a perfect and standard set of security evaluation index system for Cloud ERP security issues. Considering the issues which may be inherent in the conventional Enterprise Resource Planning systems and new ones brought by Cloud Computing, Cloud Platform, Cloud ERP, we use text mining technology to mine the related standard files in this paper. Since the standard literature is based on the comprehensive results of technology, science and practice, its characteristics of standardization, objectivity and validity also ensure the effectiveness and applicability of indicators. By using NLP methods, we construct a Cloud ERP security evaluation index system consisting of five parts: evaluation method security, information access security, data security, management security and others. The entropy method is used to assign weights for the Cloud ERP security evaluation index system, which is more scientific and rational and the method moderates the lack of objectivity caused by traditional evaluation methods which over-relies on evaluators. By analyzing cases of Cloud ERP collected in various application reports and research papers, we improve our system and find that compared to other evaluation systems, our system can evaluate the security of Cloud ERP more comprehensively.

Objectivity (philosophy)

10.3233/faia200733

Cite

Citations (0)

A Laser Plane Attitude Evaluation Method for Rail Profile Measurement Sensors

Sensors (2023)

Le Wang Hao Wang Qiang Han Yue Fang Shengchun Wang

The non-coplanar lasers on both sides of the rail during full-section rail profile measurement based on line-structured light vision will cause the measured profile to be distorted, resulting in measurement errors. Currently, in the field of rail profile measurement, there are no effective methods for evaluating laser plane attitude, and it is impossible to determine the degree of laser coplanarity quantitatively and accurately. This study proposes an evaluation method based on fitting planes in response to this problem. Real-time fitting of laser planes with three planar targets of different heights provides information about the laser plane attitude on both sides of the rails. On this basis, laser coplanarity evaluation criteria were developed to determine whether the laser planes on both sides of the rails are coplanar. Using the method in this study, the laser plane attitude can be quantified and accurately assessed on both sides, effectively resolving the problem with traditional methods that can only assess the laser plane attitude qualitatively and roughly, thereby providing a solid foundation for calibration and error correction of the measurement system.

Coplanarity

Line (geometry)

Basis (linear algebra)

10.3390/s23104586

Cite

Citations (3)

Hypergraph expressing low-rank feature selection algorithm

Multimedia Tools and Applications (2017)

Yue Fang Yangding Li Cong Lei Yonggang Li Xuelian Deng

Hypergraph

Feature (linguistics)

Rank (graph theory)

10.1007/s11042-017-5235-3

Cite

Citations (1)

Research on the Construction of Cloud ERP Ecosystem Security Evaluation Index System

Frontiers in artificial intelligence and applications (2020)

Zhu-Jun Wang Qin Su Yue Fang

Cloud ERP system realizes the functions of ERP in the way of cloud service, which greatly improves the service level of ERP and reduces the cost for enterprise to use ERP. Cloud ERP system has the characteristics of open ecology, therefore the interactions between users increase, which brings data security risks to the cloud ERP platform. This paper discusses the security of cloud ERP system from the perspective of ecosystem, and puts forward the concept and connotation of cloud ERP ecosystem. The LDA method is used to analyze the literature related to the security of ecosystem and cloud ERP open ecology, based on which seven main themes are summarized. It mainly focuses on system risk assessment, information security management and platform ecosystem construction. Based on LDA analysis results, this paper constructs a cloud ERP ecosystem security evaluation system using expert interviews from three dimensions: developer ecosystem security, user ecosystem security and device-system-user ecosystem security.

Connotation

10.3233/faia200735

Cite

Citations (0)

Low-Rank Sparse Subspace for Spectral Clustering

IEEE Transactions on Knowledge and Data Engineering (2018)

Xiaofeng Zhu Shichao Zhang Yonggang Li Jilian Zhang Lifeng Yang

Traditional graph clustering methods consist of two sequential steps, i.e., constructing an affinity matrix from the original data and then performing spectral clustering on the resulting affinity matrix. This two-step strategy achieves optimal solution for each step separately, but cannot guarantee that it will obtain the globally optimal clustering results. Moreover, the affinity matrix directly learned from the original data will seriously affect the clustering performance, since high-dimensional data are usually noisy and may contain redundancy. To address the above issues, this paper proposes a Low-rank Sparse Subspace (LSS) clustering method via dynamically learning the affinity matrix from low-dimensional space of the original data. Specifically, we learn a transformation matrix to project the original data to their low-dimensional space, by conducting feature selection and subspace learning in the sample self-representation framework. Then, we utilize the rank constraint and the affinity matrix directly obtained from the original data to construct a dynamic and intrinsic affinity matrix. Moreover, each of these three matrices is updated iteratively while fixing the other two. In this way, the affinity matrix learned from the low-dimensional space is the final clustering results. Extensive experiments are conducted on both synthetic and real datasets to show that our proposed LSS method outperforms the state-of-the-art clustering methods.

Spectral Clustering

Clustering high-dimensional data

Rank (graph theory)

Matrix (chemical analysis)

Constrained clustering

Affinity propagation

10.1109/tkde.2018.2858782

Cite

Citations (184)

Unsupervised feature selection by combining subspace learning with feature self-representation

Pattern Recognition Letters (2017)

Yangding Li Cong Lei Yue Fang Rongyao Hu Yonggang Li

Coefficient matrix

Representation

Feature (linguistics)

Regularization

Feature Learning

10.1016/j.patrec.2017.09.022

Cite

Citations (20)

Generalization of Electric Vehicles' Battery Cell Models with Machine Learning

Yue Fang

Battery cells are the core of any electric vehicles (EV). It is important to get theaccurate value of different characteristics in a battery cell for safety and lifetime concern.And Two machine lea ...

Battery electric vehicle

Source

Cite

Citations (0)