International Academy of Ecology and Environmental Sciences Network Pharmacology 2410-1000 1 4 2016 12 1 A mathematical model for dynamics of occurrence probability of missing links in predicted missing link list 86-94 EN WenJun Zhang http://www.iaees.org/publications/journals/np/articles/2016-1(4)/model-for-dynamics-of-occurrence-probability-of-missing-links.pdf> 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.