Presentation 2019-07-22
An Efficient Graph Sequence Clustering by Merging Vertices and Deleting Edges
Yoshitaka Kimura, Akihiro Inokuchi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we tackle the problem for clustering a graph sequence to discover the change of community in social networks. Recently, we have proposed a method for clustering graph sequences based on the spectral clustering. In the method, we succeeded to reduce the calculation time by merging vertices in the graph sequence in its preprocessing stage. However, the method is not suitable for graph sequences consisting of dense graphs. In this paper, we propose a method to improve the accuracy while maintaining the computational efficiency by deleting some edges with small weights and creating another graph sequence consisting of sparse graphs in the preprocessing stage.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) spectral clustering, / graph sequence / social networks / data mining
Paper # AI2019-5
Date of Issue 2019-07-15 (AI)

Conference Information
Committee AI
Conference Date 2019/7/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Naoki Fukuta(Shizuoka Univ.)
Vice Chair Yuichi Sei(Univ. of Electro-Comm.) / Yuko Sakurai(AIST)
Secretary Yuichi Sei(Osaka Univ.) / Yuko Sakurai(Tokyo Univ. of Agriculture and Technology)
Assistant

Paper Information
Registration To Technical Committee on Artificial Intelligence and Knowledge-Based Processing
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Efficient Graph Sequence Clustering by Merging Vertices and Deleting Edges
Sub Title (in English)
Keyword(1) spectral clustering,
Keyword(2) graph sequence
Keyword(3) social networks
Keyword(4) data mining
1st Author's Name Yoshitaka Kimura
1st Author's Affiliation Kwansei Gakuin University(Kwansei Gakuin Univ.)
2nd Author's Name Akihiro Inokuchi
2nd Author's Affiliation Kwansei Gakuin University(Kwansei Gakuin Univ.)
Date 2019-07-22
Paper # AI2019-5
Volume (vol) vol.119
Number (no) AI-139
Page pp.pp.21-26(AI),
#Pages 6
Date of Issue 2019-07-15 (AI)