Home

Selforganizology, 2015, 2(4): 91-101
[XML] [EndNote] [RefManager] [BibTex] [ Full PDF (142K)][Supplementary Material (4K)] [Comment/Review Article]

Article

Prediction of missing connections in the network: A node-similarity based algorithm

WenJun Zhang
School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, China; International Academy of Ecology and Environmental Sciences, Hong Kong

Received 18 August 2015;Accepted 21 September 2015;Published online 1 December 2015
IAEES

Abstract
In present study, I proposed a node-similarity based algorithm for prediction of missing connections in the network. In this algorithm, whether a node vk can connect to vi or not, depending on the similarity between vk and vi, the similarities between vi and its adjacent nodes, the similarities between vk and the adjacent nodes of vi, and the degree of node vi, and vice versa. Pearson correlation measure, cosine measure, and (negative) Euclidean distance measure (the three measures are for interval attributes), contingency correlation measure (for nominal attributes), and Jaccard coefficient measure (for binary attributes) were used as the between-node similarity. Two application examples showed a better prediction of the algorithm (approximately 60% of missing connections are successfully predicted). Matlab codes of the algorithm were provided.

Keywords network;connections;prediction;node similarity;algorithm.


International Academy of Ecology and Environmental Sciences. E-mail: office@iaees.org
Copyright © 2009-2024 International Academy of Ecology and Environmental Sciences. All rights reserved.
Web administrator: office@iaees.org, website@iaees.org; Last modified: 2024/5/9


Translate page to: