Результаты исследований: Публикации в книгах, отчётах, сборниках, трудах конференций › статья в сборнике материалов конференции › научная › Рецензирование
Generalized optimization framework for graph-based semi-supervised learning. / Avrachenkov, Konstantin; Gonçalves, Paulo; Mishenin, Alexey; Sokol, Marina.
Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012. Society for Industrial and Applied Mathematics, 2012. стр. 966-974 (Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012).Результаты исследований: Публикации в книгах, отчётах, сборниках, трудах конференций › статья в сборнике материалов конференции › научная › Рецензирование
}
TY - GEN
T1 - Generalized optimization framework for graph-based semi-supervised learning
AU - Avrachenkov, Konstantin
AU - Gonçalves, Paulo
AU - Mishenin, Alexey
AU - Sokol, Marina
PY - 2012
Y1 - 2012
N2 - We develop a generalized optimization framework for graphbased semi-supervised learning. The framework gives as particular cases the Standard Laplacian, Normalized Laplacian and PageRank based methods. We have also provided new probabilistic interpretation based on random walks and characterized the limiting behaviour of the methods. The random walk based interpretation allows us to explain differences between the performances of methods with different smoothing kernels. It appears that the PageRank based method is robust with respect to the choice of the regularization parameter and the labelled data. We illustrate our theoretical results with two realistic datasets, characterizing different challenges: Les Miserables characters social network and Wikipedia hyper-link graph. The graphbased semi-supervised learning classifies the Wikipedia articles with very good precision and perfect recall employing only the information about the hyper-text links.
AB - We develop a generalized optimization framework for graphbased semi-supervised learning. The framework gives as particular cases the Standard Laplacian, Normalized Laplacian and PageRank based methods. We have also provided new probabilistic interpretation based on random walks and characterized the limiting behaviour of the methods. The random walk based interpretation allows us to explain differences between the performances of methods with different smoothing kernels. It appears that the PageRank based method is robust with respect to the choice of the regularization parameter and the labelled data. We illustrate our theoretical results with two realistic datasets, characterizing different challenges: Les Miserables characters social network and Wikipedia hyper-link graph. The graphbased semi-supervised learning classifies the Wikipedia articles with very good precision and perfect recall employing only the information about the hyper-text links.
KW - Graph-based semi-supervised learning
KW - Optimization framework
KW - Pagerank
KW - Wikipedia article classifi- cation
UR - http://www.scopus.com/inward/record.url?scp=84880220875&partnerID=8YFLogxK
U2 - 10.1137/1.9781611972825.83
DO - 10.1137/1.9781611972825.83
M3 - Conference contribution
AN - SCOPUS:84880220875
SN - 9781611972320
T3 - Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012
SP - 966
EP - 974
BT - Proceedings of the 12th SIAM International Conference on Data Mining, SDM 2012
PB - Society for Industrial and Applied Mathematics
T2 - 12th SIAM International Conference on Data Mining, SDM 2012
Y2 - 26 April 2012 through 28 April 2012
ER -
ID: 98680839