HetPathMine: A Novel Transductive Classification Algorithm on Heterogeneous Information Networks

Chen Luo, Renchu Guan, Zhe Wang, Chenghua Lin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

20 Citations (Scopus)

Abstract

Transductive classification (TC) using a small labeled data to help classifying all the unlabeled data in information networks. It is an important data mining task on information networks. Various classification methods have been proposed for this task. However, most of these methods are proposed for homogeneous networks but not for heterogeneous ones, which include multi-typed objects and relations and may contain more useful semantic information. In this paper, we firstly use the concept of meta path to represent the different relation paths in heterogeneous networks and propose a novel meta path selection model. Then we extend the transductive classification problem to heterogeneous information networks and propose a novel algorithm, named HetPathMine. The experimental results show that: (1) HetPathMine can get higher accuracy than the existing transductive classification methods and (2) the weight obtained by HetPathMine for each meta path is consistent with human intuition or real-world situations.
Original languageEnglish
Title of host publicationAdvances in Information Retrieval
Subtitle of host publication36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings
EditorsMaarten de Rijke, Tom Kenter, Arjen P. de Vries, ChengXiang Zhai, Franciska de Jong, Kira Radinsky, Katja Hofman
PublisherSpringer
Pages210-221
Number of pages12
ISBN (Electronic)978-3-319-06028-6
ISBN (Print)978-3-319-06027-9
DOIs
Publication statusPublished - 2014

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume 8416
ISSN (Electronic)0302-9743

Fingerprint

Heterogeneous networks
Data mining
Semantics

Cite this

Luo, C., Guan, R., Wang, Z., & Lin, C. (2014). HetPathMine: A Novel Transductive Classification Algorithm on Heterogeneous Information Networks. In M. de Rijke, T. Kenter, A. P. de Vries, C. Zhai, F. de Jong, K. Radinsky, & K. Hofman (Eds.), Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings (pp. 210-221). (Lecture Notes in Computer Science; Vol. 8416 ). Springer . https://doi.org/10.1007/978-3-319-06028-6_18

HetPathMine : A Novel Transductive Classification Algorithm on Heterogeneous Information Networks. / Luo, Chen; Guan, Renchu; Wang, Zhe; Lin, Chenghua.

Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings. ed. / Maarten de Rijke; Tom Kenter; Arjen P. de Vries; ChengXiang Zhai; Franciska de Jong; Kira Radinsky; Katja Hofman. Springer , 2014. p. 210-221 (Lecture Notes in Computer Science; Vol. 8416 ).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Luo, C, Guan, R, Wang, Z & Lin, C 2014, HetPathMine: A Novel Transductive Classification Algorithm on Heterogeneous Information Networks. in M de Rijke, T Kenter, AP de Vries, C Zhai, F de Jong, K Radinsky & K Hofman (eds), Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings. Lecture Notes in Computer Science, vol. 8416 , Springer , pp. 210-221. https://doi.org/10.1007/978-3-319-06028-6_18
Luo C, Guan R, Wang Z, Lin C. HetPathMine: A Novel Transductive Classification Algorithm on Heterogeneous Information Networks. In de Rijke M, Kenter T, de Vries AP, Zhai C, de Jong F, Radinsky K, Hofman K, editors, Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings. Springer . 2014. p. 210-221. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-06028-6_18
Luo, Chen ; Guan, Renchu ; Wang, Zhe ; Lin, Chenghua. / HetPathMine : A Novel Transductive Classification Algorithm on Heterogeneous Information Networks. Advances in Information Retrieval: 36th European Conference on IR Research, ECIR 2014, Amsterdam, The Netherlands, April 13-16, 2014. Proceedings. editor / Maarten de Rijke ; Tom Kenter ; Arjen P. de Vries ; ChengXiang Zhai ; Franciska de Jong ; Kira Radinsky ; Katja Hofman. Springer , 2014. pp. 210-221 (Lecture Notes in Computer Science).
@inproceedings{38234eddc03345c6b21909a292606f78,
title = "HetPathMine: A Novel Transductive Classification Algorithm on Heterogeneous Information Networks",
abstract = "Transductive classification (TC) using a small labeled data to help classifying all the unlabeled data in information networks. It is an important data mining task on information networks. Various classification methods have been proposed for this task. However, most of these methods are proposed for homogeneous networks but not for heterogeneous ones, which include multi-typed objects and relations and may contain more useful semantic information. In this paper, we firstly use the concept of meta path to represent the different relation paths in heterogeneous networks and propose a novel meta path selection model. Then we extend the transductive classification problem to heterogeneous information networks and propose a novel algorithm, named HetPathMine. The experimental results show that: (1) HetPathMine can get higher accuracy than the existing transductive classification methods and (2) the weight obtained by HetPathMine for each meta path is consistent with human intuition or real-world situations.",
author = "Chen Luo and Renchu Guan and Zhe Wang and Chenghua Lin",
year = "2014",
doi = "10.1007/978-3-319-06028-6_18",
language = "English",
isbn = "978-3-319-06027-9",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "210--221",
editor = "{de Rijke}, Maarten and Tom Kenter and {de Vries}, {Arjen P. } and ChengXiang Zhai and {de Jong}, Franciska and Kira Radinsky and Katja Hofman",
booktitle = "Advances in Information Retrieval",

}

TY - GEN

T1 - HetPathMine

T2 - A Novel Transductive Classification Algorithm on Heterogeneous Information Networks

AU - Luo, Chen

AU - Guan, Renchu

AU - Wang, Zhe

AU - Lin, Chenghua

PY - 2014

Y1 - 2014

N2 - Transductive classification (TC) using a small labeled data to help classifying all the unlabeled data in information networks. It is an important data mining task on information networks. Various classification methods have been proposed for this task. However, most of these methods are proposed for homogeneous networks but not for heterogeneous ones, which include multi-typed objects and relations and may contain more useful semantic information. In this paper, we firstly use the concept of meta path to represent the different relation paths in heterogeneous networks and propose a novel meta path selection model. Then we extend the transductive classification problem to heterogeneous information networks and propose a novel algorithm, named HetPathMine. The experimental results show that: (1) HetPathMine can get higher accuracy than the existing transductive classification methods and (2) the weight obtained by HetPathMine for each meta path is consistent with human intuition or real-world situations.

AB - Transductive classification (TC) using a small labeled data to help classifying all the unlabeled data in information networks. It is an important data mining task on information networks. Various classification methods have been proposed for this task. However, most of these methods are proposed for homogeneous networks but not for heterogeneous ones, which include multi-typed objects and relations and may contain more useful semantic information. In this paper, we firstly use the concept of meta path to represent the different relation paths in heterogeneous networks and propose a novel meta path selection model. Then we extend the transductive classification problem to heterogeneous information networks and propose a novel algorithm, named HetPathMine. The experimental results show that: (1) HetPathMine can get higher accuracy than the existing transductive classification methods and (2) the weight obtained by HetPathMine for each meta path is consistent with human intuition or real-world situations.

U2 - 10.1007/978-3-319-06028-6_18

DO - 10.1007/978-3-319-06028-6_18

M3 - Conference contribution

SN - 978-3-319-06027-9

T3 - Lecture Notes in Computer Science

SP - 210

EP - 221

BT - Advances in Information Retrieval

A2 - de Rijke, Maarten

A2 - Kenter, Tom

A2 - de Vries, Arjen P.

A2 - Zhai, ChengXiang

A2 - de Jong, Franciska

A2 - Radinsky, Kira

A2 - Hofman, Katja

PB - Springer

ER -