Home

Network Pharmacology, 2016, 1(4): 86-94
[XML] [EndNote] [RefManager] [BibTex] [ Full PDF (147K)][Supplementary Material (647K)] [Comment/Review Article]

Article

A mathematical model for dynamics of occurrence probability of missing links in predicted missing link list

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

Received 25 December 2015;Accepted 15 February 2016;Published 1 December 2016
IAEES

Abstract
In most of the link prediction methods, all predicted missing links are ranked according to their scores. In the practical application of prediction results, starting from the first link that has the highest score in the ranking list, we verify each link one by one through experiments or other ways. Nevertheless, how to find an occurrence pattern of true missing links in the ranking list has seldomly reported. In present study, I proposed a mathematical model for relationship between cumulative number of predicted true missing links (y) and cumulative number of predicted missing links (x): y=K(1-e-rx/K), where K is the expected total number of true missing links, and r is the intrinsic (maximum) occurrence probability of true missing links. It can be used to predict the changes of occurrence probability of true missing links, assess the effectiveness of a prediction method, and help find the mechanism of link missing in the network. The model was validated by six prediction methods using the data of tumor pathways.

Keywords mathematical model;missing links;prediction;occurrence probability;tumor pathways.


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: 2024/5/6


Translate page to: