A comparison of collective classification techniques on network data

Ozge Ataseven; Yusuf Yaslan

doi:10.23919/CISTI.2017.7976061

A comparison of collective classification techniques on network data

Ozge Ataseven, Yusuf Yaslan

Department of Computer Engineering

Istanbul Technical University

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

Abstract

Collective Classification techniques aim to improve the classification performance of linked data by utilizing unknown nodes in the network that are classified by using known nodes and network structure. In this paper, we consider both single and multi-labeled linked data classification problem using local and global classification algorithms. Initially, single-labeled linked data classification problem is evaluated using ICA-KNN, ICA-Naïve Bayes, LBP and MF algorithms on bibliographic datasets. Then we extend our experiments on terrorism relation multi-labeled linked dataset by using ML-LBP, ML-MF global classification algorithms. The experimental results show that for single-labeled linked data the best classification accuracy is obtained by MF global classification algorithm. For multi-labeled data both ML-LBP and ML-MF algorithms perform similarly.

Original language	English
Title of host publication	Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017
Editors	Luis Paulo Reis, Alvaro Rocha, Braulio Alturas, Carlos Costa, Manuel Perez Cota
Publisher	IEEE Computer Society
ISBN (Electronic)	9789899843479
DOIs	https://doi.org/10.23919/CISTI.2017.7976061
Publication status	Published - 11 Jul 2017
Event	12th Iberian Conference on Information Systems and Technologies, CISTI 2017 - Lisbon, Portugal Duration: 21 Jun 2017 → 24 Jun 2017

Publication series

Name	Iberian Conference on Information Systems and Technologies, CISTI
ISSN (Print)	2166-0727
ISSN (Electronic)	2166-0735

Conference

Conference	12th Iberian Conference on Information Systems and Technologies, CISTI 2017
Country/Territory	Portugal
City	Lisbon
Period	21/06/17 → 24/06/17

Bibliographical note

Publisher Copyright:
© 2017 AISTI.

Keywords

Bibliographic dataset
Collective classification
Multi-labeled classification
Single-labeled classification

Access to Document

10.23919/CISTI.2017.7976061

Cite this

Ataseven, O., & Yaslan, Y. (2017). A comparison of collective classification techniques on network data. In L. P. Reis, A. Rocha, B. Alturas, C. Costa, & M. P. Cota (Eds.), Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017 Article 7976061 (Iberian Conference on Information Systems and Technologies, CISTI). IEEE Computer Society. https://doi.org/10.23919/CISTI.2017.7976061

Ataseven, Ozge ; Yaslan, Yusuf. / A comparison of collective classification techniques on network data. Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017. editor / Luis Paulo Reis ; Alvaro Rocha ; Braulio Alturas ; Carlos Costa ; Manuel Perez Cota. IEEE Computer Society, 2017. (Iberian Conference on Information Systems and Technologies, CISTI).

@inproceedings{076a8c0ddb0a481684356a851c2e991b,

title = "A comparison of collective classification techniques on network data",

abstract = "Collective Classification techniques aim to improve the classification performance of linked data by utilizing unknown nodes in the network that are classified by using known nodes and network structure. In this paper, we consider both single and multi-labeled linked data classification problem using local and global classification algorithms. Initially, single-labeled linked data classification problem is evaluated using ICA-KNN, ICA-Na{\"i}ve Bayes, LBP and MF algorithms on bibliographic datasets. Then we extend our experiments on terrorism relation multi-labeled linked dataset by using ML-LBP, ML-MF global classification algorithms. The experimental results show that for single-labeled linked data the best classification accuracy is obtained by MF global classification algorithm. For multi-labeled data both ML-LBP and ML-MF algorithms perform similarly.",

keywords = "Bibliographic dataset, Collective classification, Multi-labeled classification, Single-labeled classification",

author = "Ozge Ataseven and Yusuf Yaslan",

note = "Publisher Copyright: {\textcopyright} 2017 AISTI.; 12th Iberian Conference on Information Systems and Technologies, CISTI 2017 ; Conference date: 21-06-2017 Through 24-06-2017",

year = "2017",

month = jul,

day = "11",

doi = "10.23919/CISTI.2017.7976061",

language = "English",

series = "Iberian Conference on Information Systems and Technologies, CISTI",

publisher = "IEEE Computer Society",

editor = "Reis, {Luis Paulo} and Alvaro Rocha and Braulio Alturas and Carlos Costa and Cota, {Manuel Perez}",

booktitle = "Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017",

address = "United States",

}

Ataseven, O & Yaslan, Y 2017, A comparison of collective classification techniques on network data. in LP Reis, A Rocha, B Alturas, C Costa & MP Cota (eds), Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017., 7976061, Iberian Conference on Information Systems and Technologies, CISTI, IEEE Computer Society, 12th Iberian Conference on Information Systems and Technologies, CISTI 2017, Lisbon, Portugal, 21/06/17. https://doi.org/10.23919/CISTI.2017.7976061

A comparison of collective classification techniques on network data. / Ataseven, Ozge; Yaslan, Yusuf.
Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017. ed. / Luis Paulo Reis; Alvaro Rocha; Braulio Alturas; Carlos Costa; Manuel Perez Cota. IEEE Computer Society, 2017. 7976061 (Iberian Conference on Information Systems and Technologies, CISTI).

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

TY - GEN

T1 - A comparison of collective classification techniques on network data

AU - Ataseven, Ozge

AU - Yaslan, Yusuf

PY - 2017/7/11

Y1 - 2017/7/11

N2 - Collective Classification techniques aim to improve the classification performance of linked data by utilizing unknown nodes in the network that are classified by using known nodes and network structure. In this paper, we consider both single and multi-labeled linked data classification problem using local and global classification algorithms. Initially, single-labeled linked data classification problem is evaluated using ICA-KNN, ICA-Naïve Bayes, LBP and MF algorithms on bibliographic datasets. Then we extend our experiments on terrorism relation multi-labeled linked dataset by using ML-LBP, ML-MF global classification algorithms. The experimental results show that for single-labeled linked data the best classification accuracy is obtained by MF global classification algorithm. For multi-labeled data both ML-LBP and ML-MF algorithms perform similarly.

AB - Collective Classification techniques aim to improve the classification performance of linked data by utilizing unknown nodes in the network that are classified by using known nodes and network structure. In this paper, we consider both single and multi-labeled linked data classification problem using local and global classification algorithms. Initially, single-labeled linked data classification problem is evaluated using ICA-KNN, ICA-Naïve Bayes, LBP and MF algorithms on bibliographic datasets. Then we extend our experiments on terrorism relation multi-labeled linked dataset by using ML-LBP, ML-MF global classification algorithms. The experimental results show that for single-labeled linked data the best classification accuracy is obtained by MF global classification algorithm. For multi-labeled data both ML-LBP and ML-MF algorithms perform similarly.

KW - Bibliographic dataset

KW - Collective classification

KW - Multi-labeled classification

KW - Single-labeled classification

UR - http://www.scopus.com/inward/record.url?scp=85027015143&partnerID=8YFLogxK

U2 - 10.23919/CISTI.2017.7976061

DO - 10.23919/CISTI.2017.7976061

M3 - Conference contribution

AN - SCOPUS:85027015143

T3 - Iberian Conference on Information Systems and Technologies, CISTI

BT - Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017

A2 - Reis, Luis Paulo

A2 - Rocha, Alvaro

A2 - Alturas, Braulio

A2 - Costa, Carlos

A2 - Cota, Manuel Perez

PB - IEEE Computer Society

T2 - 12th Iberian Conference on Information Systems and Technologies, CISTI 2017

Y2 - 21 June 2017 through 24 June 2017

ER -

Ataseven O, Yaslan Y. A comparison of collective classification techniques on network data. In Reis LP, Rocha A, Alturas B, Costa C, Cota MP, editors, Atas da 12a Conferencia Iberica de Sistemas e Tecnologias de Informacao, CISTI 2017 / Proceedings of the 12th Iberian Conference on Information Systems and Technologies, CISTI 2017. IEEE Computer Society. 2017. 7976061. (Iberian Conference on Information Systems and Technologies, CISTI). doi: 10.23919/CISTI.2017.7976061

A comparison of collective classification techniques on network data

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this