Graph-based semi-supervised learning as a generative model

Jingrui He; Jaime Carbonell; Yan Liu

Graph-based semi-supervised learning as a generative model

Jingrui He, Jaime Carbonell, Yan Liu

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper proposes and develops a new graph-based semi-supervised learning method. Different from previous graph-based methods that are based on discriminative models, our method is essentially a generative model in that the class conditional probabilities are estimated by graph propagation and the class priors are estimated by linear regression. Experimental results on various datasets show that the proposed method is superior to existing graph-based semi-supervised learning methods, especially when the labeled subset alone proves insufficient to estimate meaningful class priors.

Original language	English (US)
Title of host publication	IJCAI International Joint Conference on Artificial Intelligence
Pages	2492-2497
Number of pages	6
State	Published - 2007
Externally published	Yes
Event	20th International Joint Conference on Artificial Intelligence, IJCAI 2007 - Hyderabad, India Duration: Jan 6 2007 → Jan 12 2007

Other

Other	20th International Joint Conference on Artificial Intelligence, IJCAI 2007
Country/Territory	India
City	Hyderabad
Period	1/6/07 → 1/12/07

ASJC Scopus subject areas

Artificial Intelligence

Cite this

@inproceedings{6c8c31145118476a85597d979d16821d,

title = "Graph-based semi-supervised learning as a generative model",

abstract = "This paper proposes and develops a new graph-based semi-supervised learning method. Different from previous graph-based methods that are based on discriminative models, our method is essentially a generative model in that the class conditional probabilities are estimated by graph propagation and the class priors are estimated by linear regression. Experimental results on various datasets show that the proposed method is superior to existing graph-based semi-supervised learning methods, especially when the labeled subset alone proves insufficient to estimate meaningful class priors.",

author = "Jingrui He and Jaime Carbonell and Yan Liu",

year = "2007",

language = "English (US)",

pages = "2492--2497",

booktitle = "IJCAI International Joint Conference on Artificial Intelligence",

note = "20th International Joint Conference on Artificial Intelligence, IJCAI 2007 ; Conference date: 06-01-2007 Through 12-01-2007",

}

TY - GEN

T1 - Graph-based semi-supervised learning as a generative model

AU - He, Jingrui

AU - Carbonell, Jaime

AU - Liu, Yan

PY - 2007

Y1 - 2007

N2 - This paper proposes and develops a new graph-based semi-supervised learning method. Different from previous graph-based methods that are based on discriminative models, our method is essentially a generative model in that the class conditional probabilities are estimated by graph propagation and the class priors are estimated by linear regression. Experimental results on various datasets show that the proposed method is superior to existing graph-based semi-supervised learning methods, especially when the labeled subset alone proves insufficient to estimate meaningful class priors.

AB - This paper proposes and develops a new graph-based semi-supervised learning method. Different from previous graph-based methods that are based on discriminative models, our method is essentially a generative model in that the class conditional probabilities are estimated by graph propagation and the class priors are estimated by linear regression. Experimental results on various datasets show that the proposed method is superior to existing graph-based semi-supervised learning methods, especially when the labeled subset alone proves insufficient to estimate meaningful class priors.

UR - http://www.scopus.com/inward/record.url?scp=84880903985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880903985&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84880903985

SP - 2492

EP - 2497

BT - IJCAI International Joint Conference on Artificial Intelligence

T2 - 20th International Joint Conference on Artificial Intelligence, IJCAI 2007

Y2 - 6 January 2007 through 12 January 2007

ER -

Graph-based semi-supervised learning as a generative model

Abstract

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this