复杂网络近优链路预测的叠加模型
Stacking Models for Nearly Optimal Link Prediction in Complex Networks
摘要：
大多数真实世界的网络都是不完全观察到的。能够准确预测哪些链路缺失的算法可以大大加快网络数据的收集速度，提高网络模型的有效性。现在存在许多算法来预测缺失的链接，给定一个部分观察到的网络，但它仍然是未知的是否存在一个单一的最佳预测，如何链接可预测性不同的方法和网络从不同的领域，以及如何接近最优性目前的方法。我们回答这些问题，系统地评估203个单独的链接预测算法，代表三个流行的家庭的方法，适用于一个大型语料库的548个结构不同的网络从六个科学领域。我们首先表明，个别算法表现出广泛的多样性的预测误差，这样，没有一个预测或家庭是最好的，或最坏的，在所有现实的输入。然后，我们通过元学习来利用这种多样性，构建一系列“堆叠”模型，将预测器组合到一个算法中。应用于广泛的合成网络，我们可以分析计算最佳性能，这些堆叠模型实现了最佳或接近最佳的准确性水平。应用于现实世界的网络，堆叠模型也是优越的，但它们的准确性因领域而异，这表明链接预测在社交网络中可能比在生物或技术网络中更容易。这些结果表明，链接预测的最新技术来自于组合各个算法，从而实现接近最佳的预测。最后，我们简要讨论了进一步改进这些结果的局限性和机会。
英文摘要：
Most real-world networks are incompletely observed. Algorithms that can accurately predict which links are missing can dramatically speedup the collection of network data and improve the validity of network models. 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 548 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 via meta-learning 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 also 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 achieves nearly optimal predictions. We close with a brief discussion of limitations and opportunities for further improvement of these results.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
