Abstract
Recent years, the studies of link prediction have been overwhelmingly emphasizing on undirected networks. Compared with it, how to identify missing and spurious interactions in directed networks has received less attention and still is not well understood. In this paper, we make use of classical link prediction indices for undirected networks, adapt them to directed version which could predict both the existence and direction of an arc between two nodes, and investigate their prediction ability on six real-world directed networks. Experimental results demonstrate that those modified indices perform quite well in directed networks. Compared with bifan predictor, some of them can provide more accurate predictions.
Highlights
Network is an effective and efficient tool to describe realworld complex systems [1, 2], such as social, biological, traffic, and information systems, where nodes represent individuals, proteins, airports, web pages, and so forth and links denote the relations and interactions between them
It is worth noting that Preferential Attachment (PA) index, which is regarded as the worst predictor in undirected networks, surprisingly performs quite well in the experiments
This paper studied how to identify both missing and spurious interactions in directed networks
Summary
Network is an effective and efficient tool to describe realworld complex systems [1, 2], such as social, biological, traffic, and information systems, where nodes represent individuals, proteins, airports, web pages, and so forth and links denote the relations and interactions between them. While making great efforts to understand the structural features and evolutionary mechanism of networks, scientists gradually realize that the inaccuracy and incompleteness of data sets is a significant obstacle to the research [3, 4]. To address this issue, link prediction algorithms have been adopted to extract the missing information, identify spurious interactions, and reconstruct network. In online social networks, very likely but not-yet-existent links can be recommended as promising friendship, which can help users in finding new friends; in biological networks, compared to blindly checking all possible protein-protein interactions, accurate prediction of the most likely existent ones can dramatically reduce the experimental cost; in ecommerce, with the help of recommendation systems, sellers enhance their sales by watching customers’ purchases and recommending other goods to them in which they may be interested [6]; in security domain, link prediction methods could be used to assist identifying groups of terrorist or criminals [7]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Distributed Sensor Networks
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.
