Efficient semi-supervised learning on locally informative multiple graphs

Motoki Shiga, Hiroshi Mamitsuka

Research output: Contribution to journalArticlepeer-review

17 Citations (Scopus)


We address an issue of semi-supervised learning on multiple graphs, over which informative subgraphs are distributed. One application under this setting can be found in molecular biology, where different types of gene networks are generated depending upon experiments. Here an important problem is to annotate unknown genes by using functionally known genes, which connect to unknown genes in gene networks, in which informative parts vary over networks. We present a powerful, time-efficient approach for this problem by combining soft spectral clustering with label propagation for multiple graphs. We demonstrate the effectiveness and efficiency of our approach using both synthetic and real biological datasets.

Original languageEnglish
Pages (from-to)1035-1049
Number of pages15
JournalPattern Recognition
Issue number3
Publication statusPublished - 2012 Mar
Externally publishedYes


  • EM (Expectation Maximization) algorithm
  • Graph integration
  • Label propagation
  • Semi-supervised learning
  • Soft spectral clustering

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence


Dive into the research topics of 'Efficient semi-supervised learning on locally informative multiple graphs'. Together they form a unique fingerprint.

Cite this