Abstract
Real world complex networks are indirect representation of complex systems. They grow over time. These networks are fragmented and raucous in practice. An important concern about complex network is link prediction. Link prediction aims to determine the possibility of probable edges. The link prediction demand is often spotted in social networks for recommending new friends, and, in recommender systems for recommending new items (movies, gadgets etc) based on earlier shopping history. In this work, we propose a new link prediction algorithm namely “Common Neighbor and Centrality based Parameterized Algorithm” (CCPA) to suggest the formation of new links in complex networks. Using AUC (Area Under the receiver operating characteristic Curve) as evaluation criterion, we perform an extensive experimental evaluation of our proposed algorithm on eight real world data sets, and against eight benchmark algorithms. The results validate the improved performance of our proposed algorithm.
Highlights
Complex networks are effective descriptions of real world networks, where real world problems can be modeled in the form of complex network graphs[1]
We found that the AUC of Common Neighbor and Distance (CND) (0.76) is very close to that of Common Neighbor and Centrality based Parameterized Algorithm (CCPA)
We observed that CCPA is the best performing algorithm on 5 data sets whereas CND is the best performing algorithm on two data sets (Karate and Polbook) only
Summary
Considerable body of literature is devoted to the study of link prediction in complex networks (see[7] and the references therein). Yang and Zhang[20], introduced an algorithm based on the common neighbors and distance metric to predict link in a variety of real world networks from the available topological structure of the network. Using AUC as evaluating criterion, authors analyzed the performance of their proposed algorithm using real world data sets. The authors further proposed a novel algorithm based on resource allocation process, which achieved superior experimental performance than common neighbor algorithm. Murata and Moriyasu[22] presented an algorithm based on the proximity measures and weights of existing links in a weighted graph to predict possible future interactions in online social networks. Our proposed algorithm is based on two vital properties of nodes, namely the number of common neighbors and their centrality. Closeness centrality, we propose a new algorithm for missing link prediction. Zachary for over three years from 1970 to 1972 to study the clash arose between instructor and administrator[28]
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 callDisclaimer: 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.
