Second-level degree-based entity resolution in online social networks

Joo Young Lee; Rasheed Hussain; Victor Rivera; Davlatbek Isroilov

doi:10.1007/s13278-018-0499-9

Second-level degree-based entity resolution in online social networks

Joo Young Lee^*, Rasheed Hussain, Victor Rivera, Davlatbek Isroilov

^*Corresponding author for this work

Department of Electrical & Electronic Engineering

Research output: Contribution to journal › Article (Academic Journal) › peer-review

4 Citations (Scopus)

Abstract

Abundance of online social platforms allows users to create more than one profile in different social networks. Several issues arise due to multiple user identities including data integration and information retrieval in social networks. Identifying user profiles across social platforms is known as entity resolution. In this paper, we propose a degree-based method to attack entity resolution problems. More precisely, we utilize the degree of users and their friends to identify user profiles. Our results show that, without help of critical information such as e-mail addresses, the proposed method can outperform existing string matching-based solutions as well as popular classifiers such as SVM and Naive Bayes.

Original language	English
Article number	19
Journal	Social Network Analysis and Mining
Volume	8
Issue number	1
DOIs	https://doi.org/10.1007/s13278-018-0499-9
Publication status	Published - 1 Dec 2018

Bibliographical note

Publisher Copyright:
© 2018, Springer-Verlag GmbH Austria, part of Springer Nature.

Keywords

Entity resolution
Social network analysis
String matching
User profile matching

Access to Document

10.1007/s13278-018-0499-9

Handle.net

Persistent link

Cite this

@article{2ff44a9e654a46e7ba62f5cfc7e4c100,

title = "Second-level degree-based entity resolution in online social networks",

abstract = "Abundance of online social platforms allows users to create more than one profile in different social networks. Several issues arise due to multiple user identities including data integration and information retrieval in social networks. Identifying user profiles across social platforms is known as entity resolution. In this paper, we propose a degree-based method to attack entity resolution problems. More precisely, we utilize the degree of users and their friends to identify user profiles. Our results show that, without help of critical information such as e-mail addresses, the proposed method can outperform existing string matching-based solutions as well as popular classifiers such as SVM and Naive Bayes.",

keywords = "Entity resolution, Social network analysis, String matching, User profile matching",

author = "Lee, {Joo Young} and Rasheed Hussain and Victor Rivera and Davlatbek Isroilov",

note = "Publisher Copyright: {\textcopyright} 2018, Springer-Verlag GmbH Austria, part of Springer Nature.",

year = "2018",

month = dec,

day = "1",

doi = "10.1007/s13278-018-0499-9",

language = "English",

volume = "8",

journal = "Social Network Analysis and Mining",

issn = "1869-5450",

publisher = "Springer Verlag",

number = "1",

}

TY - JOUR

T1 - Second-level degree-based entity resolution in online social networks

AU - Lee, Joo Young

AU - Hussain, Rasheed

AU - Rivera, Victor

AU - Isroilov, Davlatbek

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Abundance of online social platforms allows users to create more than one profile in different social networks. Several issues arise due to multiple user identities including data integration and information retrieval in social networks. Identifying user profiles across social platforms is known as entity resolution. In this paper, we propose a degree-based method to attack entity resolution problems. More precisely, we utilize the degree of users and their friends to identify user profiles. Our results show that, without help of critical information such as e-mail addresses, the proposed method can outperform existing string matching-based solutions as well as popular classifiers such as SVM and Naive Bayes.

AB - Abundance of online social platforms allows users to create more than one profile in different social networks. Several issues arise due to multiple user identities including data integration and information retrieval in social networks. Identifying user profiles across social platforms is known as entity resolution. In this paper, we propose a degree-based method to attack entity resolution problems. More precisely, we utilize the degree of users and their friends to identify user profiles. Our results show that, without help of critical information such as e-mail addresses, the proposed method can outperform existing string matching-based solutions as well as popular classifiers such as SVM and Naive Bayes.

KW - Entity resolution

KW - Social network analysis

KW - String matching

KW - User profile matching

UR - http://www.scopus.com/inward/record.url?scp=85044156766&partnerID=8YFLogxK

U2 - 10.1007/s13278-018-0499-9

DO - 10.1007/s13278-018-0499-9

M3 - Article (Academic Journal)

AN - SCOPUS:85044156766

SN - 1869-5450

VL - 8

JO - Social Network Analysis and Mining

JF - Social Network Analysis and Mining

IS - 1

M1 - 19

ER -

Second-level degree-based entity resolution in online social networks

Abstract

Bibliographical note

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this