Link prediction for tree-like networks

Ke Ke Shang, Tong Chen Li, Michael Small, David Burton, Yan Wang

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Link prediction is the problem of predicting the location of either unknown or fake links from uncertain structural information of a network. Link prediction algorithms are useful in gaining insight into different network structures from partial observations of exemplars. However, existing link prediction algorithms only focus on regular complex networks and are overly dependent on either the closed triangular structure of networks or the so-called preferential attachment phenomenon. The performance of these algorithms on highly sparse or treelike networks is poor. In this letter, we proposed a method that is based on the network heterogeneity. We test our algorithms for three real large sparse networks: a metropolitan water distribution network, a Twitter network, and a sexual contact network. We find that our method is effective and performs better than traditional algorithms, especially for the Twitter network. We further argue that heterogeneity is the most obvious defining pattern for complex networks, while other statistical properties failed to be predicted. Moreover, preferential attachment based link prediction performed poorly and hence we infer that preferential attachment is not a plausible model for the genesis of many networks. We also suggest that heterogeneity is an important mechanism for online information propagation.

Original languageEnglish
Article number061103
JournalChaos
Volume29
Issue number6
DOIs
Publication statusPublished - 1 Jun 2019

Fingerprint

Trees (mathematics)
Prediction
predictions
Complex networks
Preferential Attachment
Complex Networks
attachment
Electric power distribution
Partial Observation
Distribution Network
Network Structure
Statistical property
Triangular
Water
Contact
Propagation
Unknown
Closed
Dependent

Cite this

Shang, K. K., Li, T. C., Small, M., Burton, D., & Wang, Y. (2019). Link prediction for tree-like networks. Chaos, 29(6), [061103]. https://doi.org/10.1063/1.5107440
Shang, Ke Ke ; Li, Tong Chen ; Small, Michael ; Burton, David ; Wang, Yan. / Link prediction for tree-like networks. In: Chaos. 2019 ; Vol. 29, No. 6.
@article{5a1c30a58ae34403a49f783b6f9d06d8,
title = "Link prediction for tree-like networks",
abstract = "Link prediction is the problem of predicting the location of either unknown or fake links from uncertain structural information of a network. Link prediction algorithms are useful in gaining insight into different network structures from partial observations of exemplars. However, existing link prediction algorithms only focus on regular complex networks and are overly dependent on either the closed triangular structure of networks or the so-called preferential attachment phenomenon. The performance of these algorithms on highly sparse or treelike networks is poor. In this letter, we proposed a method that is based on the network heterogeneity. We test our algorithms for three real large sparse networks: a metropolitan water distribution network, a Twitter network, and a sexual contact network. We find that our method is effective and performs better than traditional algorithms, especially for the Twitter network. We further argue that heterogeneity is the most obvious defining pattern for complex networks, while other statistical properties failed to be predicted. Moreover, preferential attachment based link prediction performed poorly and hence we infer that preferential attachment is not a plausible model for the genesis of many networks. We also suggest that heterogeneity is an important mechanism for online information propagation.",
author = "Shang, {Ke Ke} and Li, {Tong Chen} and Michael Small and David Burton and Yan Wang",
year = "2019",
month = "6",
day = "1",
doi = "10.1063/1.5107440",
language = "English",
volume = "29",
journal = "Chaos",
issn = "1054-1500",
publisher = "ACOUSTICAL SOC AMER AMER INST PHYSICS",
number = "6",

}

Shang, KK, Li, TC, Small, M, Burton, D & Wang, Y 2019, 'Link prediction for tree-like networks' Chaos, vol. 29, no. 6, 061103. https://doi.org/10.1063/1.5107440

Link prediction for tree-like networks. / Shang, Ke Ke; Li, Tong Chen; Small, Michael; Burton, David; Wang, Yan.

In: Chaos, Vol. 29, No. 6, 061103, 01.06.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Link prediction for tree-like networks

AU - Shang, Ke Ke

AU - Li, Tong Chen

AU - Small, Michael

AU - Burton, David

AU - Wang, Yan

PY - 2019/6/1

Y1 - 2019/6/1

N2 - Link prediction is the problem of predicting the location of either unknown or fake links from uncertain structural information of a network. Link prediction algorithms are useful in gaining insight into different network structures from partial observations of exemplars. However, existing link prediction algorithms only focus on regular complex networks and are overly dependent on either the closed triangular structure of networks or the so-called preferential attachment phenomenon. The performance of these algorithms on highly sparse or treelike networks is poor. In this letter, we proposed a method that is based on the network heterogeneity. We test our algorithms for three real large sparse networks: a metropolitan water distribution network, a Twitter network, and a sexual contact network. We find that our method is effective and performs better than traditional algorithms, especially for the Twitter network. We further argue that heterogeneity is the most obvious defining pattern for complex networks, while other statistical properties failed to be predicted. Moreover, preferential attachment based link prediction performed poorly and hence we infer that preferential attachment is not a plausible model for the genesis of many networks. We also suggest that heterogeneity is an important mechanism for online information propagation.

AB - Link prediction is the problem of predicting the location of either unknown or fake links from uncertain structural information of a network. Link prediction algorithms are useful in gaining insight into different network structures from partial observations of exemplars. However, existing link prediction algorithms only focus on regular complex networks and are overly dependent on either the closed triangular structure of networks or the so-called preferential attachment phenomenon. The performance of these algorithms on highly sparse or treelike networks is poor. In this letter, we proposed a method that is based on the network heterogeneity. We test our algorithms for three real large sparse networks: a metropolitan water distribution network, a Twitter network, and a sexual contact network. We find that our method is effective and performs better than traditional algorithms, especially for the Twitter network. We further argue that heterogeneity is the most obvious defining pattern for complex networks, while other statistical properties failed to be predicted. Moreover, preferential attachment based link prediction performed poorly and hence we infer that preferential attachment is not a plausible model for the genesis of many networks. We also suggest that heterogeneity is an important mechanism for online information propagation.

UR - http://www.scopus.com/inward/record.url?scp=85068172347&partnerID=8YFLogxK

U2 - 10.1063/1.5107440

DO - 10.1063/1.5107440

M3 - Article

VL - 29

JO - Chaos

JF - Chaos

SN - 1054-1500

IS - 6

M1 - 061103

ER -

Shang KK, Li TC, Small M, Burton D, Wang Y. Link prediction for tree-like networks. Chaos. 2019 Jun 1;29(6). 061103. https://doi.org/10.1063/1.5107440