Name
Abstract | ||
|---|---|---|
Social network link prediction has gained significant attention and become a key research focus over the last two decades. The prediction of missing links in the current network and emerging or broken links in future networks is essential for the understanding of their evolutionary nature. Social networks are changing dynamically over time. Link inference in dynamic social networks is an extremely challenging process and few link prediction methods consider their evolving nature. The aim of this paper is to comprehensively review, analyze, discuss and evaluate state-of-the-art link prediction methods in dynamic social networks. The leading link prediction methods and techniques that network science has produced are categorized and compared. Features and evaluation metrics for each method are presented. Finally, some future directions and recommendations are provided. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/CIST.2018.8596511 | 2018 IEEE 5th International Congress on Information Science and Technology (CiSt) |
Keywords | Field | DocType |
link prediction,link inference,dynamic social etworks,evolutionary | Network science,Social network,Inference,Computer science,Prediction algorithms,Artificial intelligence,Probabilistic logic,Machine learning | Conference |
ISSN | ISBN | Citations |
2327-185X | 978-1-5386-4386-0 | 0 |
PageRank | References | Authors |
0.34 | 18 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Mohammad Marjan | 1 | 0 | 0.34 |
| Nazar Zaki | 2 | 139 | 14.31 |
| Elfadil A. Mohamed | 3 | 0 | 0.34 |