scholarly journals Graph Convolutional Networks using Heat Kernel for Semi-supervised Learning

2019 ◽  
Author(s):  
Bingbing Xu ◽  
Huawei Shen ◽  
Qi Cao ◽  
Keting Cen ◽  
Xueqi Cheng
Keyword(s):  
Heat Kernel ◽  
Supervised Learning ◽  
Low Frequency ◽  
Weighted Average ◽  
Relevant Information ◽  
Semisupervised Learning ◽  
Heat Diffusion ◽  
Graph Structure ◽  
Convolutional Networks ◽  
Benchmark Datasets

Graph convolutional networks gain remarkable success in semi-supervised learning on graph-structured data. The key to graph-based semisupervised learning is capturing the smoothness of labels or features over nodes exerted by graph structure. Previous methods, spectral methods and spatial methods, devote to defining graph convolution as a weighted average over neighboring nodes, and then learn graph convolution kernels to leverage the smoothness to improve the performance of graph-based semi-supervised learning. One open challenge is how to determine appropriate neighborhood that reflects relevant information of smoothness manifested in graph structure. In this paper, we propose GraphHeat, leveraging heat kernel to enhance low-frequency filters and enforce smoothness in the signal variation on the graph. GraphHeat leverages the local structure of target node under heat diffusion to determine its neighboring nodes flexibly, without the constraint of order suffered by previous methods. GraphHeat achieves state-of-the-art results in the task of graph-based semi-supervised classification across three benchmark datasets: Cora, Citeseer and Pubmed.

scholarly journals Adaptive Graph Convolution Using Heat Kernel for Attributed Graph Clustering

2020 ◽  
Vol 10 (4) ◽  
pp. 1473
Author(s):  
Danyang Zhu ◽  
Shudong Chen ◽  
Xiuhui Ma ◽  
Rong Du
Keyword(s):  
Heat Kernel ◽  
Selection Criterion ◽  
Graph Clustering ◽  
Relevant Information ◽  
Heat Diffusion ◽  
Structural Relationships ◽  
Attributed Graph ◽  
Attributed Graphs ◽  
Cluster Distance ◽  
Benchmark Datasets

Attributed graphs contain a lot of node features and structural relationships, and how to utilize their inherent information sufficiently to improve graph clustering performance has attracted much attention. Although existing advanced methods exploit graph convolution to capture the global structure of an attributed graph and achieve obvious improvements for clustering results, they cannot determine the optimal neighborhood that reflects the relevant information of connected nodes in a graph. To address this limitation, we propose a novel adaptive graph convolution using a heat kernel model for attributed graph clustering (AGCHK), which exploits the similarity among nodes under heat diffusion to flexibly restrict the neighborhood of the center node and enforce the graph smoothness. Additionally, we take the Davies–Bouldin index (DBI) instead of the intra-cluster distance individually as the selection criterion to adaptively determine the order of graph convolution. The clustering results of AGCHK on three benchmark datasets—Cora, Citeseer, and Pubmed—are all more than 1% higher than the current advanced model AGC, and 12% on the Wiki dataset especially, which obtains a state-of-the-art result in the task of attributed graph clustering.

Despeckling of Sar Images Using Shrinkage of Two-Dimensional Discrete Orthonormal S-Transform

2020 ◽  
pp. 2150023
Author(s):  
Priya R. Kamath ◽  
Kedarnath Senapati ◽  
P. Jidesh
Keyword(s):  
High Frequency ◽  
Low Frequency ◽  
Relevant Information ◽  
Detection Algorithm ◽  
Two Dimensional ◽  
Sar Images ◽  
S Transform ◽  
Processing Step ◽  
Frequency Components ◽  
Shock Filter

Speckles are inherent to SAR. They hide and undermine several relevant information contained in the SAR images. In this paper, a despeckling algorithm using the shrinkage of two-dimensional discrete orthonormal S-transform (2D-DOST) coefficients in the transform domain along with shock filter is proposed. Also, an attempt has been made as a post-processing step to preserve the edges and other details while removing the speckle. The proposed strategy involves decomposing the SAR image into low and high-frequency components and processing them separately. A shock filter is used to smooth out the small variations in low-frequency components, and the high-frequency components are treated with a shrinkage of 2D-DOST coefficients. The edges, for enhancement, are detected using a ratio-based edge detection algorithm. The proposed method is tested, verified, and compared with some well-known models on C-band and X-band SAR images. A detailed experimental analysis is illustrated.

scholarly journals Discovering latent node Information by graph attention network

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Weiwei Gu ◽  
Fei Gao ◽  
Xiaodan Lou ◽  
Jiang Zhang
Keyword(s):  
Supervised Learning ◽  
Link Prediction ◽  
Learning Algorithm ◽  
Network Visualization ◽  
Graph Structure ◽  
High Quality ◽  
Node Centrality ◽  
Influential Nodes ◽  
Node Classification ◽  
Highly Cited

AbstractIn this paper, we propose graph attention based network representation (GANR) which utilizes the graph attention architecture and takes graph structure as the supervised learning information. Compared with node classification based representations, GANR can be used to learn representation for any given graph. GANR is not only capable of learning high quality node representations that achieve a competitive performance on link prediction, network visualization and node classification but it can also extract meaningful attention weights that can be applied in node centrality measuring task. GANR can identify the leading venture capital investors, discover highly cited papers and find the most influential nodes in Susceptible Infected Recovered Model. We conclude that link structures in graphs are not limited on predicting linkage itself, it is capable of revealing latent node information in an unsupervised way once a appropriate learning algorithm, like GANR, is provided.

scholarly journals Non-Intrusive Load Disaggregation Based on a Multi-Scale Attention Residual Network

2020 ◽  
Vol 10 (24) ◽  
pp. 9132
Author(s):  
Liguo Weng ◽  
Xiaodong Zhang ◽  
Junhao Qian ◽  
Min Xia ◽  
Yiqing Xu ◽  
...  
Keyword(s):  
Smart Grids ◽  
Recognition Rate ◽  
Low Frequency ◽  
Attention Mechanism ◽  
Learning Ability ◽  
Residual Network ◽  
Multi Scale ◽  
Energy Disaggregation ◽  
Benchmark Datasets ◽  
Load Disaggregation

Non-intrusive load disaggregation (NILD) is of great significance to the development of smart grids. Current energy disaggregation methods extract features from sequences, and this process easily leads to a loss of load features and difficulties in detecting, resulting in a low recognition rate of low-use electrical appliances. To solve this problem, a non-intrusive sequential energy disaggregation method based on a multi-scale attention residual network is proposed. Multi-scale convolutions are used to learn features, and the attention mechanism is used to enhance the learning ability of load features. The residual learning further improves the performance of the algorithm, avoids network degradation, and improves the precision of load decomposition. The experimental results on two benchmark datasets show that the proposed algorithm has more advantages than the existing algorithms in terms of load disaggregation accuracy and judgments of the on/off state, and the attention mechanism can further improve the disaggregation accuracy of low-frequency electrical appliances.

Semi-supervised Intrusive Appliance Load Monitoring in Smart Energy Monitoring System

2021 ◽  
Vol 17 (3) ◽  
pp. 1-20
Author(s):  
Vanh Khuyen Nguyen ◽  
Wei Emma Zhang ◽  
Adnan Mahmood
Keyword(s):  
Supervised Learning ◽  
Monitoring System ◽  
End Users ◽  
Training Data ◽  
Smart Device ◽  
Energy Monitoring ◽  
Load Monitoring ◽  
Benchmark Datasets ◽  
Smart Home Environment ◽  
Smart Appliance

Intrusive Load Monitoring (ILM) is a method to measure and collect the energy consumption data of individual appliances via smart plugs or smart sockets. A major challenge of ILM is automatic appliance identification, in which the system is able to determine automatically a label of the active appliance connected to the smart device. Existing ILM techniques depend on labels input by end-users and are usually under the supervised learning scheme. However, in reality, end-users labeling is laboriously rendering insufficient training data to fit the supervised learning models. In this work, we propose a semi-supervised learning (SSL) method that leverages rich signals from the unlabeled dataset and jointly learns the classification loss for the labeled dataset and the consistency training loss for unlabeled dataset. The samples fit into consistency learning are generated by a transformation that is built upon weighted versions of DTW Barycenter Averaging algorithm. The work is inspired by two recent advanced works in SSL in computer vision and combines the advantages of the two. We evaluate our method on the dataset collected from our developed Internet-of-Things based energy monitoring system in a smart home environment. We also examine the method’s performances on 10 benchmark datasets. As a result, the proposed method outperforms other methods on our smart appliance datasets and most of the benchmarks datasets, while it shows competitive results on the rest datasets.

DEPSO With DTCWT Algorithm for Multimodal Medical Image Fusion

2021 ◽  
Vol 12 (4) ◽  
pp. 78-97
Author(s):  
Hassiba Talbi ◽  
Mohamed-Khireddine Kholladi
Keyword(s):  
Wavelet Transform ◽  
Image Fusion ◽  
High Frequency ◽  
Medical Image ◽  
Low Frequency ◽  
Weighted Average ◽  
Optimization Method ◽  
Absolute Maximum ◽  
Complex Wavelet Transform ◽  
Medical Image Fusion

In this paper, the authors propose an algorithm of hybrid particle swarm with differential evolution (DE) operator, termed DEPSO, with the help of a multi-resolution transform named dual tree complex wavelet transform (DTCWT) to solve the problem of multimodal medical image fusion. This hybridizing approach aims to combine algorithms in a judicious manner, where the resulting algorithm will contain the positive features of these different algorithms. This new algorithm decomposes the source images into high-frequency and low-frequency coefficients by the DTCWT, then adopts the absolute maximum method to fuse high-frequency coefficients; the low-frequency coefficients are fused by a weighted average method while the weights are estimated and enhanced by an optimization method to gain optimal results. The authors demonstrate by the experiments that this algorithm, besides its simplicity, provides a robust and efficient way to fuse multimodal medical images compared to existing wavelet transform-based image fusion algorithms.

scholarly journals Multimodal Medical Image Fusion using Guided Filter in NSCT Domain

2018 ◽  
Vol 11 (4) ◽  
pp. 1937-1946
Author(s):  
Nancy Mehta ◽  
Sumit Budhiraja
Keyword(s):  
Image Fusion ◽  
Medical Image ◽  
Similarity Index ◽  
Low Frequency ◽  
Structural Similarity ◽  
Relevant Information ◽  
Phase Congruency ◽  
Edge Preserving ◽  
Medical Image Fusion ◽  
Fused Image

Multimodal medical image fusion aims at minimizing the redundancy and collecting the relevant information using the input images acquired from different medical sensors. The main goal is to produce a single fused image having more information and has higher efficiency for medical applications. In this paper modified fusion method has been proposed in which NSCT decomposition is used to decompose the wavelet coefficients obtained after wavelet decomposition. NSCT being multidirectional,shift invariant transform provide better results.Guided filter has been used for the fusion of high frequency coefficients on account of its edge preserving property. Phase congruency is used for the fusion of low frequency coefficients due to its insensitivity to illumination contrast hence making it suitable for medical images. The simulated results show that the proposed technique shows better performance in terms of entropy, structural similarity index, Piella metric. The fusion response of the proposed technique is also compared with other fusion approaches; proving the effectiveness of the obtained fusion results.

Predicting Secondary School Students' Performance Utilizing a Semi-supervised Learning Approach

2018 ◽  
Vol 57 (2) ◽  
pp. 448-470 ◽  
Author(s):  
Ioannis E. Livieris ◽  
Konstantina Drakopoulou ◽  
Vassilis T. Tampakas ◽  
Tassos A. Mikropoulos ◽  
Panagiotis Pintelas
Keyword(s):  
Supervised Learning ◽  
Prediction Models ◽  
Learning Algorithms ◽  
Educational Data Mining ◽  
Semisupervised Learning ◽  
Research Field ◽  
Machine Learning Techniques ◽  
School Students ◽  
Supervised Learning Algorithms ◽  
And Performance

Educational data mining constitutes a recent research field which gained popularity over the last decade because of its ability to monitor students' academic performance and predict future progression. Numerous machine learning techniques and especially supervised learning algorithms have been applied to develop accurate models to predict student's characteristics which induce their behavior and performance. In this work, we examine and evaluate the effectiveness of two wrapper methods for semisupervised learning algorithms for predicting the students' performance in the final examinations. Our preliminary numerical experiments indicate that the advantage of semisupervised methods is that the classification accuracy can be significantly improved by utilizing a few labeled and many unlabeled data for developing reliable prediction models.

scholarly journals Encoding Time Series as Multi-Scale Signed Recurrence Plots for Classification Using Fully Convolutional Networks

Sensors ◽  
2020 ◽  
Vol 20 (14) ◽  
pp. 3818
Author(s):  
Ye Zhang ◽  
Yi Hou ◽  
Shilin Zhou ◽  
Kewei Ouyang
Keyword(s):  
Time Series ◽  
State Of The Art ◽  
Recurrence Plot ◽  
The State ◽  
Recurrence Plots ◽  
Convolutional Networks ◽  
Multi Scale ◽  
Fully Convolutional Networks ◽  
Benchmark Datasets ◽  
Visualization Evaluation

Recent advances in time series classification (TSC) have exploited deep neural networks (DNN) to improve the performance. One promising approach encodes time series as recurrence plot (RP) images for the sake of leveraging the state-of-the-art DNN to achieve accuracy. Such an approach has been shown to achieve impressive results, raising the interest of the community in it. However, it remains unsolved how to handle not only the variability in the distinctive region scale and the length of sequences but also the tendency confusion problem. In this paper, we tackle the problem using Multi-scale Signed Recurrence Plots (MS-RP), an improvement of RP, and propose a novel method based on MS-RP images and Fully Convolutional Networks (FCN) for TSC. This method first introduces phase space dimension and time delay embedding of RP to produce multi-scale RP images; then, with the use of asymmetrical structure, constructed RP images can represent very long sequences (>700 points). Next, MS-RP images are obtained by multiplying designed sign masks in order to remove the tendency confusion. Finally, FCN is trained with MS-RP images to perform classification. Experimental results on 45 benchmark datasets demonstrate that our method improves the state-of-the-art in terms of classification accuracy and visualization evaluation.

Sign in / Sign up

Export Citation Format

Share Document