Link Prediction in Social Networks using Vertex Entropy

Many link prediction methods have been put out and tested on several actual networks. The weights of linkages are rarely considered in these studies. Taking both the network's structure and link weight into account is required for link prediction. Previous researchers mostly overlooked the topo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of recent technology and engineering 2023-07, Vol.12 (2), p.102-108
Hauptverfasser: Shubham, Kumar, Dr. Rajeev, Chauhan, Dr. Naveen
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Many link prediction methods have been put out and tested on several actual networks. The weights of linkages are rarely considered in these studies. Taking both the network's structure and link weight into account is required for link prediction. Previous researchers mostly overlooked the topological structure data in favour of the naturally occurring link weight. With the use of the concept of entropy, a new link prediction algorithm has been put forth in this paper.When used in real-time social networks, this algorithm outperforms the industry standard techniques. This paper concentrated on both topological structural information which focuses on calculating the vertex entropy of each very vertex and link weight in the proposed method. Both weighted and unweighted networks can benefit from the proposed method. Unipartite and bipartite networks can also use the suggested methods. Further, results demonstrate that the proposed method performs better than competing or traditional strategies, particularly when targeted social networks are sufficiently dense.
ISSN:2277-3878
2277-3878
DOI:10.35940/ijrte.A7593.0712223