Graph Learning: A Survey

Xia, Feng; Sun, Ke; Yu, Shuo; Aziz, Abdul; Wan, Liangtian; Pan, Shirui; Liu, Huan

doi:10.1109/tai.2021.3076021

Cited by 260 publications

(89 citation statements)

References 134 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1. 1) Laplacian KSRC (LapKSRC) with L 2,1 norm: Based on the kernel sparse representation-based classifier method, the Laplacian regularization term [48], [49] is added in order to better preserve the association between DNA sequences.…”

Section: B Multi-view Learning Modelmentioning

confidence: 99%

Identification of DNA N4-methylcytosine Sites via Multiview Kernel Sparse Representation Model

Tiwari

Yang

et al. 2023

IEEE Trans. Artif. Intell.

View full text Add to dashboard Cite

Identifying DNA N4-methylcytosine (4mC) sites is 1 of great significance in biological research, such as chromatin 2 structure, DNA stability, DNA-protein interaction and controlling 3 gene expression. However, the traditional sequencing technology to 4 identify 4mC sites is very time-consuming. In order to detect 4mC 5 sites, we develop a multi-view learning method for achieving more 6 effectively via merging multiple feature spaces. Furthermore, we 7 think about whether the multi-view learning method can improve 8 the across species classification ability by fusing data of multiple 9 species. In our study, we propose a multi-view Laplacian kernel 10 sparse representation-based classifier, called MvLapKSRC-HSIC. 11 First, we make use of three feature extraction methods (PSTNP, 12 NCP, DPP) to extract the DNA sequence features. MvLapKSRC-13 HSIC uses a kernel sparse representation-based classifier with 14 graph regularization. In order to maintain the independence 15 between various views, we add a multi-view regularization term 16 constructed by Hilbert-Schmidt independence criterion (HSIC).17In the experiments, MvLapKSRC-HSIC is applied on six datasets, 18 so as to compare with other popular methods in single species 19 and cross-species experiments. All experimental results show that 20 MvLapKSRC-HSIC is superior to other outstanding methods on 21 both single species and cross-species. Importantly, MvLapKSRC-22 HSIC can identify a series of potential DNA 4mC sites, which 23 have not yet been experimentally evaluate on multiple species 24 and merit further research. 25 Impact Statement-As an important DNA modification, 4mC 26 plays an important role in DNA replication, expression and other 27 biological processes, so the detection of DNA 4mC sites has always 28 been a research hotspot. In view of the fact that previous studies 29 have neglected the importance of different prior knowledge may 30 be different, they have not chosen the appropriate methods to fuse 31 prior knowledge. Here, we propose a method to predict DNA 4mC 32 sites based on multi-view learning, which effectively integrates 33 different prior knowledge. On the six benchmark data sets, the

show abstract

Section: B Multi-view Learning Modelmentioning

confidence: 99%

Identification of DNA N4-methylcytosine Sites via Multiview Kernel Sparse Representation Model

Tiwari

Yang

et al. 2023

IEEE Trans. Artif. Intell.

View full text Add to dashboard Cite

show abstract

“…Currently, a large number of researchers have concentrated their efforts on such an emerging area with the technology of big data, yielding impressive results (Bai et al, 2020;Hou et al, 2020;Zhang et al, 2020;Xia et al, 2021). A systematic review is urgently needed to sort out the results and challenges of current research and to provide references for educational policymaking and subsequent research.…”

Section: Introductionmentioning

confidence: 99%

Educational Anomaly Analytics: Features, Methods, and Challenges

et al. 2022

Self Cite

View full text Add to dashboard Cite

Anomalies in education affect the personal careers of students and universities' retention rates. Understanding the laws behind educational anomalies promotes the development of individual students and improves the overall quality of education. However, the inaccessibility of educational data hinders the development of the field. Previous research in this field used questionnaires, which are time- and cost-consuming and hardly applicable to large-scale student cohorts. With the popularity of educational management systems and the rise of online education during the prevalence of COVID-19, a large amount of educational data is available online and offline, providing an unprecedented opportunity to explore educational anomalies from a data-driven perspective. As an emerging field, educational anomaly analytics rapidly attracts scholars from a variety of fields, including education, psychology, sociology, and computer science. This paper intends to provide a comprehensive review of data-driven analytics of educational anomalies from a methodological standpoint. We focus on the following five types of research that received the most attention: course failure prediction, dropout prediction, mental health problems detection, prediction of difficulty in graduation, and prediction of difficulty in employment. Then, we discuss the challenges of current related research. This study aims to provide references for educational policymaking while promoting the development of educational anomaly analytics as a growing field.

show abstract

“…In the era of big data, graph learning [1] has attracted considerable attention owing to its wide applications in data science, machine learning, and image processing, etc. In the process of dealing with problems related to graph learning, such as graph networks [2,3], image processing [4,5], and reinforcement learning [6], it is often necessary to solve the eigenproblem of the Laplacian matrix of a fully connected graph to avoid dropping crucial nonlocal information.…”

Section: Introductionmentioning

confidence: 99%

A quantum algorithm for solving eigenproblem of the Laplacian matrix of a fully connected weighted graph

Liu¹,

Qin²,

Wan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Solving eigenproblem of the Laplacian matrix of a fully connected graph has wide applications in data science, machine learning, and image processing, etc. However, it is very challenging because it involves expensive matrix operations. Here, we propose an efficient quantum algorithm to solve it based on a general assumption that the element of each vertex and its norms can be accessed via a quantum random access data structure. Specifically, we propose in detail how to construct the block-encodings of operators containing the information of the weight matrix and the degree matrix respectively, and further obtain the block-encoding of the Laplacian matrix. This enables us to implement the quantum simulation of the Laplacian matrix by employing the optimal Hamiltonian simulation technique based on the block-encoding framework. Afterwards, the eigenvalues and eigenvectors of the Laplacian matrix are extracted by the quantum phase estimation algorithm. Compared with its classical counterpart, our algorithm has a polynomial speedup on the number of vertices and an exponential speedup on the dimension of each vertex over its classical counterpart. We also show that our algorithm can be extended to solve the eigenproblem of symmetric (nonsymmetric) normalized Laplacian matrix.

show abstract

Graph Learning: A Survey

Cited by 260 publications

References 134 publications

Identification of DNA N4-methylcytosine Sites via Multiview Kernel Sparse Representation Model

Identification of DNA N4-methylcytosine Sites via Multiview Kernel Sparse Representation Model

Educational Anomaly Analytics: Features, Methods, and Challenges

A quantum algorithm for solving eigenproblem of the Laplacian matrix of a fully connected weighted graph

Contact Info

Product

Resources

About