Abstract
Various structural features of networks have been applied to develop link prediction methods. However, because different features highlight different aspects of network structural properties, it is very difficult to benefit from all of the features that might be available. In this paper, we investigate the role of network topology in predicting missing links from the perspective of information theory. In this way, the contributions of different structural features to link prediction are measured in terms of their values of information. Then, an information-theoretic model is proposed that is applicable to multiple structural features. Furthermore, we design a novel link prediction index, called Neighbor Set Information (NSI), based on the information-theoretic model. According to our experimental results, the NSI index performs well in real-world networks, compared with other typical proximity indices.
Highlights
Neighbors (CN) is a basic index based on local network structural properties but has a relatively high prediction accuracy[15]
We develop an information-theoretic model for link prediction, which is applicable to various structural features
We develop an information-theoretic model that treats the link prediction problem as the evaluation of the uncertainty that a link exists
Summary
Neighbors (CN) is a basic index based on local network structural properties but has a relatively high prediction accuracy[15]. Liu et al recently proposed a Fast Blocking probabilistic Model based on a greedy strategy, which can reduce the computation complexity and improve the prediction accuracy[17]. In this model, link likelihoods are estimated by considering link densities within and among communities. In the Mutual Information index, the feature of common neighbors is considered to facilitate prediction and the link likelihood of a node pair is denoted as the conditional self-information of the event that the node pair is connected when their common neighbors are given. We design a novel link prediction index called Neighbor Set Information (NSI), which uses two types of local structural features. We test the NSI index in twelve real-world networks and find that it performs well compared with other structure based indices
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.