Abstract
Most real-world networks are incompletely observed. Algorithms that can accurately predict which links are missing can dramatically speed up network data collection and improve network model validation. Many algorithms now exist for predicting missing links, given a partially observed network, but it has remained unknown whether a single best predictor exists, how link predictability varies across methods and networks from different domains, and how close to optimality current methods are. We answer these questions by systematically evaluating 203 individual link predictor algorithms, representing three popular families of methods, applied to a large corpus of 550 structurally diverse networks from six scientific domains. We first show that individual algorithms exhibit a broad diversity of prediction errors, such that no one predictor or family is best, or worst, across all realistic inputs. We then exploit this diversity using network-based metalearning to construct a series of “stacked” models that combine predictors into a single algorithm. Applied to a broad range of synthetic networks, for which we may analytically calculate optimal performance, these stacked models achieve optimal or nearly optimal levels of accuracy. Applied to real-world networks, stacked models are superior, but their accuracy varies strongly by domain, suggesting that link prediction may be fundamentally easier in social networks than in biological or technological networks. These results indicate that the state of the art for link prediction comes from combining individual algorithms, which can achieve nearly optimal predictions. We close with a brief discussion of limitations and opportunities for further improvements.
Funder
National Science Foundation
Army Research Office
Publisher
Proceedings of the National Academy of Sciences
Reference38 articles.
1. Effects of missing data in social networks;Kossinets;Soc. Network.,2006
2. Computationally efficient link prediction in a variety of social networks;Fire;ACM Trans. Intell. Syst. Technol. (TIST),2013
3. Link prediction in complex networks: A survey;Lü;Phys. Stat. Mech. Appl. A,2011
4. M. Nagarajan , “Predicting future scientific discoveries based on a networked analysis of the past literature” in Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (Association for Computing Machinery, New York, NY, 2015), pp. 2019–2028.
5. What’s different about social media networks? A framework and research agenda;Kane;MIS Q.,2014
Cited by
89 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献