Exploratory factor analysis of graphical features for link prediction in social networks

Lale Madahali, Lotfi Najjar, Margeret Hall

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

Social networks attract much attention due to their ability to replicate social interactions at scale. Link prediction, or the assessment of which unconnected nodes are likely to connect in the future, is an interesting but non-trivial research area. Three approaches exist to deal with the link-prediction problem: feature-based models, Bayesian probabilistic models, and probabilistic relational models. In feature-based methods, graphical features are extracted and used for classification. Usually, these features are subdivided into three feature groups based on their formula. Some formulas are extracted based on neighborhood graph traverse. Accordingly, there exist three groups of features: neighborhood features, path-based features, and node-based features. In this paper, we attempt to validate the underlying structure of topological features used in feature-based link prediction. The results of our analysis indicate differing results from the prevailing grouping of these features, which indicates that current literatures’ classification of feature groups should be redefined. Thus, the contribution of this work is exploring the factor loading of graphical features in link prediction in social networks. To the best of our knowledge, no prior studies had addressed it.

Original languageEnglish (US)
Title of host publicationSpringer Proceedings in Complexity
PublisherSpringer
Pages17-31
Number of pages15
DOIs
StatePublished - Jan 1 2019

Publication series

NameSpringer Proceedings in Complexity
ISSN (Print)2213-8684
ISSN (Electronic)2213-8692

Fingerprint

Exploratory Factor Analysis
Factor analysis
Social Networks
Prediction
Probabilistic Model
Graphical Methods
Relational Model
Social Interaction
Bayesian Model
Vertex of a graph
Grouping
Likely
Path
Graphics
Graph in graph theory

Keywords

  • Data mining
  • Exploratory factor analysis
  • Social networks analysis

ASJC Scopus subject areas

  • Applied Mathematics
  • Modeling and Simulation
  • Computer Science Applications

Cite this

Madahali, L., Najjar, L., & Hall, M. (2019). Exploratory factor analysis of graphical features for link prediction in social networks. In Springer Proceedings in Complexity (pp. 17-31). (Springer Proceedings in Complexity). Springer. https://doi.org/10.1007/978-3-030-14459-3_2

Exploratory factor analysis of graphical features for link prediction in social networks. / Madahali, Lale; Najjar, Lotfi; Hall, Margeret.

Springer Proceedings in Complexity. Springer, 2019. p. 17-31 (Springer Proceedings in Complexity).

Research output: Chapter in Book/Report/Conference proceedingChapter

Madahali, L, Najjar, L & Hall, M 2019, Exploratory factor analysis of graphical features for link prediction in social networks. in Springer Proceedings in Complexity. Springer Proceedings in Complexity, Springer, pp. 17-31. https://doi.org/10.1007/978-3-030-14459-3_2
Madahali L, Najjar L, Hall M. Exploratory factor analysis of graphical features for link prediction in social networks. In Springer Proceedings in Complexity. Springer. 2019. p. 17-31. (Springer Proceedings in Complexity). https://doi.org/10.1007/978-3-030-14459-3_2
Madahali, Lale ; Najjar, Lotfi ; Hall, Margeret. / Exploratory factor analysis of graphical features for link prediction in social networks. Springer Proceedings in Complexity. Springer, 2019. pp. 17-31 (Springer Proceedings in Complexity).
@inbook{aa6611172a2d429ca9d530eacd1f3f1b,
title = "Exploratory factor analysis of graphical features for link prediction in social networks",
abstract = "Social networks attract much attention due to their ability to replicate social interactions at scale. Link prediction, or the assessment of which unconnected nodes are likely to connect in the future, is an interesting but non-trivial research area. Three approaches exist to deal with the link-prediction problem: feature-based models, Bayesian probabilistic models, and probabilistic relational models. In feature-based methods, graphical features are extracted and used for classification. Usually, these features are subdivided into three feature groups based on their formula. Some formulas are extracted based on neighborhood graph traverse. Accordingly, there exist three groups of features: neighborhood features, path-based features, and node-based features. In this paper, we attempt to validate the underlying structure of topological features used in feature-based link prediction. The results of our analysis indicate differing results from the prevailing grouping of these features, which indicates that current literatures’ classification of feature groups should be redefined. Thus, the contribution of this work is exploring the factor loading of graphical features in link prediction in social networks. To the best of our knowledge, no prior studies had addressed it.",
keywords = "Data mining, Exploratory factor analysis, Social networks analysis",
author = "Lale Madahali and Lotfi Najjar and Margeret Hall",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-14459-3_2",
language = "English (US)",
series = "Springer Proceedings in Complexity",
publisher = "Springer",
pages = "17--31",
booktitle = "Springer Proceedings in Complexity",

}

TY - CHAP

T1 - Exploratory factor analysis of graphical features for link prediction in social networks

AU - Madahali, Lale

AU - Najjar, Lotfi

AU - Hall, Margeret

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Social networks attract much attention due to their ability to replicate social interactions at scale. Link prediction, or the assessment of which unconnected nodes are likely to connect in the future, is an interesting but non-trivial research area. Three approaches exist to deal with the link-prediction problem: feature-based models, Bayesian probabilistic models, and probabilistic relational models. In feature-based methods, graphical features are extracted and used for classification. Usually, these features are subdivided into three feature groups based on their formula. Some formulas are extracted based on neighborhood graph traverse. Accordingly, there exist three groups of features: neighborhood features, path-based features, and node-based features. In this paper, we attempt to validate the underlying structure of topological features used in feature-based link prediction. The results of our analysis indicate differing results from the prevailing grouping of these features, which indicates that current literatures’ classification of feature groups should be redefined. Thus, the contribution of this work is exploring the factor loading of graphical features in link prediction in social networks. To the best of our knowledge, no prior studies had addressed it.

AB - Social networks attract much attention due to their ability to replicate social interactions at scale. Link prediction, or the assessment of which unconnected nodes are likely to connect in the future, is an interesting but non-trivial research area. Three approaches exist to deal with the link-prediction problem: feature-based models, Bayesian probabilistic models, and probabilistic relational models. In feature-based methods, graphical features are extracted and used for classification. Usually, these features are subdivided into three feature groups based on their formula. Some formulas are extracted based on neighborhood graph traverse. Accordingly, there exist three groups of features: neighborhood features, path-based features, and node-based features. In this paper, we attempt to validate the underlying structure of topological features used in feature-based link prediction. The results of our analysis indicate differing results from the prevailing grouping of these features, which indicates that current literatures’ classification of feature groups should be redefined. Thus, the contribution of this work is exploring the factor loading of graphical features in link prediction in social networks. To the best of our knowledge, no prior studies had addressed it.

KW - Data mining

KW - Exploratory factor analysis

KW - Social networks analysis

UR - http://www.scopus.com/inward/record.url?scp=85063196093&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063196093&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-14459-3_2

DO - 10.1007/978-3-030-14459-3_2

M3 - Chapter

AN - SCOPUS:85063196093

T3 - Springer Proceedings in Complexity

SP - 17

EP - 31

BT - Springer Proceedings in Complexity

PB - Springer

ER -