Measuring Large-Scale Dynamic Graph Similarity by RICom: RWR with Intergraph Compression

Lee, Jaekoo; Kim, Gunn; Yoon, Sungroh

doi:10.1109/icdm.2015.123

Cited by 2 publications

(16 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The graph representation is a powerful tool since it enables the intuitive expression of highly complex structures that have interactive elements [2]- [4], [6], [10]. Graph analysis has seen significant development as deep learning has become more sophisticated.…”

Section: Background a Learning On Graphmentioning

confidence: 99%

“…For the non-Euclidean space, the distance can be measured by computing various measures that are associated with a node. The node-level features include followings: affinity score [4], [10]; direction of nodes (in/out); centrality measures; eigenvector [30]; closeness [31]; local clustering coefficient [32]; radius [33], degree assortativity; and roles [34]. With various types of distance measurement, a threshold value is established to detect anomaly behavior.…”

Section: B Anomaly Detection On Graphmentioning

confidence: 99%

“…Traditionally, the analysis of data expressed in a graph has heavily relied on algebra, e.g., random walk [4], [7]- [10] and…”

Section: Introductionmentioning

confidence: 99%

“…The proposed deep learning model takes a graph that dynamically transforms over time or events as input and is trained using real-world datasets. One of the notable characteristics of a dynamic graph is that the properties change gradually rather than abruptly [10], as shown in Figure 1, which is involved in road networks. This observation contributed significantly to the motivation of our research.…”

Section: Introductionmentioning

confidence: 99%

“…In a dynamic graph that has scale-free graph characteristics, the prediction of affinity score vector could be used to quantify the overall changes of the graph indirectly from the perspective of an important node, e.g., a hub. Since an affinity score quantifies the structure of graph with respect to a specific node, it has been used as a major node proximity indicator in graph analysis such as cross-graph analytics, graph classification, link perdition, and anomaly detection in graphs [4], [7]- [10].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Anomaly Detection by Learning Dynamics From a Graph

2020

Self Cite

View full text Add to dashboard Cite

There exist relations, which vary with time or by an event, between high dimensional elements in most real-world datasets. A dynamic graph or network has been used as one of the remarkable approaches to represent and analyze them. In spite of the advantages of representing data in the form of graphs, it is difficult to apply representation (deep) learning to graphs. Recently, AlphaFold by DeepMind has shown remarkable results in applying deep learning to graphs. This research is part of the current effort to extend the input domain of deep learning to arbitrarily graphs and their dynamics of variations. In this paper, we propose a method to predict the evolution of graphs by learning spatio-temporal features called dynamics. The method involves two main processes: extracting spatial features from static graphs obtained at different times and learning temporal features from the time-varying connection structure. Instead of predicting the overall changes of a highly complex graph, we detect the dynamic anomaly by predicting the affinity score with respect to a node (e.g., a hub as an important factor) of a dynamics graph. This facilitates the learning dynamics of graphs having sparsity of connections by alleviating the curse of dimensions using the fact that most graphs of real-world problems are scale-free. To justify our approach, we apply our method to real-world problems such as computer networks and public transportation. Experimental results show that our approach is competitive with other existing methods.

show abstract

Section: Background a Learning On Graphmentioning

confidence: 99%

Section: B Anomaly Detection On Graphmentioning

confidence: 99%

“…Traditionally, the analysis of data expressed in a graph has heavily relied on algebra, e.g., random walk [4], [7]- [10] and…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Anomaly Detection by Learning Dynamics From a Graph

2020

Self Cite

View full text Add to dashboard Cite

show abstract

Transfer Learning for Deep Learning on Graph-Structured Data

Lee

Kim

Lee

et al. 2017

AAAI

Self Cite

View full text Add to dashboard Cite

Graphs provide a powerful means for representing complex interactions between entities. Recently, new deep learning approaches have emerged for representing and modeling graph-structured data while the conventional deep learning methods, such as convolutional neural networks and recurrent neural networks, have mainly focused on the grid-structured inputs of image and audio. Leveraged by representation learning capabilities, deep learning-based techniques can detect structural characteristics of graphs, giving promising results for graph applications. In this paper, we attempt to advance deep learning for graph-structured data by incorporating another component: transfer learning. By transferring the intrinsic geometric information learned in the source domain, our approach can construct a model for a new but related task in the target domain without collecting new data and without training a new model from scratch. We thoroughly tested our approach with large-scale real-world text data and confirmed the effectiveness of the proposed transfer learning framework for deep learning on graphs. According to our experiments, transfer learning is most effective when the source and target domains bear a high level of structural similarity in their graph representations.

show abstract

Measuring Large-Scale Dynamic Graph Similarity by RICom: RWR with Intergraph Compression

Cited by 2 publications

References 11 publications

Anomaly Detection by Learning Dynamics From a Graph

Anomaly Detection by Learning Dynamics From a Graph

Transfer Learning for Deep Learning on Graph-Structured Data

Contact Info

Product

Resources

About