Towards Co-Evolution of Random-Walk-Based Embedding and Label Propagation for Node Classification

2020 ◽  
Author(s):  
Ruixuan Zhao ◽  
Shenghang Liu ◽  
Wei Rao ◽  
Hao Xu ◽  
Chongning Na ◽  
...  
Keyword(s):  
Random Walk ◽  
Label Propagation ◽  
Node Classification

Random Walks On Biomedical Networks

2020 ◽  
Vol 17 ◽  
Author(s):  
Guiyang Zhang ◽  
Pan Wang ◽  
You Li ◽  
Guohua Huang
Keyword(s):  
Random Walk ◽  
Link Prediction ◽  
Drug Target ◽  
Target Identification ◽  
Point Of View ◽  
Random Walk With Restart ◽  
Function Annotation ◽  
Disease Biomarker ◽  
Drug Target Identification ◽  
Node Classification

Abstract: The biomedical network is becoming a fundamental tool to represent sophisticated bio-systems, while random walk models on it are becoming a sharp sword to address such challenging issues as gene function annotation, drug target identification, and disease biomarker recognition. Recently, numerous random walk models have been proposed and applied to biomedical networks. Due to good performances, the random walk is increasingly attracting more and more attention from multiple communities. In this survey, we firstly introduced various random walk models, with emphasis on the Pag-eRank and the random walk with restart. We then summarized applications of the RW on the biomedical networks from the graph learning point of view, which mainly included node classification, link prediction, cluster/community detection, and learning representation of the node. We discussed briefly its limitation and existing issues also

An improved label propagation algorithm based on node importance and random walk for community detection

2017 ◽  
Vol 31 (14) ◽  
pp. 1750162 ◽  
Author(s):  
Tianren Ma ◽  
Zhengyou Xia
Keyword(s):  
Random Walk ◽  
Community Detection ◽  
High Efficiency ◽  
Linear Time ◽  
Structural Characteristics ◽  
Rapid Development ◽  
Time Algorithm ◽  
Label Propagation ◽  
Node Importance ◽  
Propagation Algorithm

Currently, with the rapid development of information technology, the electronic media for social communication is becoming more and more popular. Discovery of communities is a very effective way to understand the properties of complex networks. However, traditional community detection algorithms consider the structural characteristics of a social organization only, with more information about nodes and edges wasted. In the meanwhile, these algorithms do not consider each node on its merits.Label propagation algorithm (LPA) is a near linear time algorithm which aims to find the community in the network. It attracts many scholars owing to its high efficiency. In recent years, there are more improved algorithms that were put forward based on LPA. In this paper, an improved LPA based on random walk and node importance (NILPA) is proposed. Firstly, a list of node importance is obtained through calculation. The nodes in the network are sorted in descending order of importance. On the basis of random walk, a matrix is constructed to measure the similarity of nodes and it avoids the random choice in the LPA. Secondly, a new metric IAS (importance and similarity) is calculated by node importance and similarity matrix, which we can use to avoid the random selection in the original LPA and improve the algorithm stability.Finally, a test in real-world and synthetic networks is given. The result shows that this algorithm has better performance than existing methods in finding community structure.

scholarly journals An Information-Explainable Random Walk Based Unsupervised Network Representation Learning Framework on Node Classification Tasks

Mathematics ◽  
2021 ◽  
Vol 9 (15) ◽  
pp. 1767
Author(s):  
Xin Xu ◽  
Yang Lu ◽  
Yupeng Zhou ◽  
Zhiguo Fu ◽  
Yanjie Fu ◽  
...  
Keyword(s):  
Random Walk ◽  
Representation Learning ◽  
Local Information ◽  
Learning Framework ◽  
Network Representation ◽  
Label Node ◽  
Label Information ◽  
Classification Tasks ◽  
Node Classification ◽  
Low Dimensional

Network representation learning aims to learn low-dimensional, compressible, and distributed representational vectors of nodes in networks. Due to the expensive costs of obtaining label information of nodes in networks, many unsupervised network representation learning methods have been proposed, where random walk strategy is one of the wildly utilized approaches. However, the existing random walk based methods have some challenges, including: 1. The insufficiency of explaining what network knowledge in the walking path-samplings; 2. The adverse effects caused by the mixture of different information in networks; 3. The poor generality of the methods with hyper-parameters on different networks. This paper proposes an information-explainable random walk based unsupervised network representation learning framework named Probabilistic Accepted Walk (PAW) to obtain network representation from the perspective of the stationary distribution of networks. In the framework, we design two stationary distributions based on nodes’ self-information and local-information of networks to guide our proposed random walk strategy to learn representational vectors of networks through sampling paths of nodes. Numerous experimental results demonstrated that the PAW could obtain more expressive representation than the other six widely used unsupervised network representation learning baselines on four real-world networks in single-label and multi-label node classification tasks.

scholarly journals Masked Graph Convolutional Network

2019 ◽  
Author(s):  
Liang Yang ◽  
Fan Wu ◽  
Yingkui Wang ◽  
Junhua Gu ◽  
Yuanfang Guo
Keyword(s):  
Supervised Classification ◽  
Label Propagation ◽  
Unstructured Data ◽  
Classification Methods ◽  
Convolutional Network ◽  
Node Attributes ◽  
Classification Tasks ◽  
Node Classification ◽  
Supervised Classification Methods ◽  
The Impact

Semi-supervised classification is a fundamental technology to process the structured and unstructured data in machine learning field. The traditional attribute-graph based semi-supervised classification methods propagate labels over the graph which is usually constructed from the data features, while the graph convolutional neural networks smooth the node attributes, i.e., propagate the attributes, over the real graph topology. In this paper, they are interpreted from the perspective of propagation, and accordingly categorized into symmetric and asymmetric propagation based methods. From the perspective of propagation, both the traditional and network based methods are propagating certain objects over the graph. However, different from the label propagation, the intuition ``the connected data samples tend to be similar in terms of the attributes", in attribute propagation is only partially valid. Therefore, a masked graph convolution network (Masked GCN) is proposed by only propagating a certain portion of the attributes to the neighbours according to a masking indicator, which is learned for each node by jointly considering the attribute distributions in local neighbourhoods and the impact on the classification results. Extensive experiments on transductive and inductive node classification tasks have demonstrated the superiority of the proposed method.

scholarly journals Multimodal person discovery using label propagation over speaking faces graphs

2019 ◽  
Author(s):  
Gabriel Barbosa Fonseca ◽  
Zenilton K. G. Patrocínio Jr ◽  
Guillaume Gravier ◽  
Silvio Jamil F. Guimarães
Keyword(s):  
Random Walk ◽  
Large Datasets ◽  
Kappa Coefficient ◽  
Label Propagation ◽  
Hierarchical Approach ◽  
Quality Of Information ◽  
Automatic Indexing ◽  
Processing Step ◽  
Propagation Methods

The indexing of large datasets is a task of great importance, since it directly impacts on the quality of information that can be retrieved from these sets. Unfortunately, some datasets are growing in size so fast that manually indexing becomes unfeasible. Automatic indexing techniques can be applied to overcome this issue, and in this study, a unsupervised technique for multimodal person discovery is proposed, which consists in detecting persons that are appearing and speaking simultaneously on a video and associating names to them. To achieve that, the data is modeled as a graph of speaking-faces, and names are extracted via OCR and propagated through the graph based on audiovisual relations between speaking faces. To propagate labels, two graph based methods are proposed, one based on random walks and the other based on a hierarchical approach. In order to assess the proposed approach, we use two graph clustering baselines, and different modality fusion approaches. On the MediaEval MPD 2017 dataset, the proposed label propagation methods outperform all literature methods except one, which uses a different approach on the pre-processing step. Even though the Kappa coefficient indicates that the random walk and the hierarchical label propagation produce highly equivalent results, the hierarchical propagation is more than 6 times faster than the random walk under same configurations.

Sign in / Sign up

Export Citation Format

Share Document