Supervised Link Prediction Using Multiple Sources

Lu, Zhengdong; Savas, Berkant; Tang, Wei; Dhillon, Inderjit S.

doi:10.1109/icdm.2010.112

Cited by 108 publications

(75 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Link prediction is the problem of predicting formation of new links in networks that evolve over time. This problem arises in applications such as friendship recommendation in social networks [26], affiliation recommendation [29], and prediction of author collaborations in scientific publications [23].…”

Section: Link Prediction In Social Networkmentioning

confidence: 99%

Parallel Clustered Low-Rank Approximation of Graphs and Its Application to Link Prediction

Sui

Lee

Whang

et al. 2013

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Social network analysis has become a major research area that has impact in diverse applications ranging from search engines to product recommendation systems. A major problem in implementing social network analysis algorithms is the sheer size of many social networks, for example, the Facebook graph has more than 900 million vertices and even small networks may have tens of millions of vertices. One solution to dealing with these large graphs is dimensionality reduction using spectral or SVD analysis of the adjacency matrix of the network, but these global techniques do not necessarily take into account local structures or clusters of the network that are critical in network analysis. A more promising approach is clustered low-rank approximation: instead of computing a global low-rank approximation, the adjacency matrix is first clustered, and then a low-rank approximation of each cluster (i.e., diagonal block) is computed. The resulting algorithm is challenging to parallelize not only because of the large size of the data sets in social network analysis, but also because it requires computing with very diverse data structures ranging from extremely sparse matrices to dense matrices. In this paper, we describe the first parallel implementation of a clustered low-rank approximation algorithm for large social network graphs, and use it to perform link prediction in parallel. Experimental results show that this implementation scales well on large distributed-memory machines; for example, on a Twitter graph with roughly 11 million vertices and 63 million edges, our implementation scales by a factor of 86 on 128 processes and takes less than 2300 seconds, while on a much larger Twitter graph with 41 million vertices and 1.2 billion edges, our implementation scales by a factor of 203 on 256 processes with a running time about 4800 seconds.

show abstract

Section: Link Prediction In Social Networkmentioning

confidence: 99%

Parallel Clustered Low-Rank Approximation of Graphs and Its Application to Link Prediction

Sui

Lee

Whang

et al. 2013

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Many of the key ideas behind network centrality measures arose out of the social sciences, where researchers were interested in understanding structural attributes of human interaction networks [13]. The ability to determine who or what is important is also valuable in many application areas, including healthcare, security, advertising, publishing and politics [2,15,21,23].…”

mentioning

confidence: 99%

Centrality Analysis for Modified Lattices

Paton¹,

Akartunalı²,

Higham³

2017

SIAM J. Matrix Anal. & Appl.

View full text Add to dashboard Cite

Abstract. We derive new, exact expressions for network centrality vectors associated with classical Watts-Strogatz style "ring plus shortcut" networks. We also derive easy-to-interpret approximations that are highly accurate in the large network limit. The analysis helps us to understand the role of the Katz parameter and the PageRank parameter, to compare linear system and eigenvalue based centrality measures, and to predict the behavior of centrality measures on more complicated networks.

show abstract

“…In addition to topological similarity indices, social relation information can also enhance link prediction, in the case of social networks [8], [21]. For general social network (such as Facebook), it is undirected.…”

Section: Social Featuresmentioning

confidence: 99%

Link Prediction in Microblog Network Using Supervised Learning with Multiple Features

Han¹,

Xu²

2016

JCP

View full text Add to dashboard Cite

Link prediction (LP) is a fundamental network analysis task. It aims to analyze the existing links and predict the missing or potential relations between users in a social network. It can help users in finding new friends, enhance their loyalties to the web sites and build a healthy social environment. In previous researches, much attention was focused on structure information or node attributes, in order to analyze the global or local properties. Considering the nature of Microblog social network, we proposed a link prediction system combining multiple features from different perspectives, and learn a classifier from these feature subsets to predict the potential links. We train classifiers using SVM, Naï ve Bayes, and Random Forest and Logistic Regression algorithms and evaluate them using the microblog network dataset. The results show that our features perform better than the traditional features, and the combination of multiple features can achieve highest accuracy.

show abstract

Supervised Link Prediction Using Multiple Sources

Cited by 108 publications

References 20 publications

Parallel Clustered Low-Rank Approximation of Graphs and Its Application to Link Prediction

Parallel Clustered Low-Rank Approximation of Graphs and Its Application to Link Prediction

Centrality Analysis for Modified Lattices

Link Prediction in Microblog Network Using Supervised Learning with Multiple Features

Contact Info

Product

Resources

About