<?xml version="1.0"?>
<metadata xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:dc="http://purl.org/dc/elements/1.1/"><dc:title>Strength prominence index</dc:title><dc:creator>Pandey,	Sakshi Dev	(Avtor)
	</dc:creator><dc:creator>Samanta,	Sovan	(Avtor)
	</dc:creator><dc:creator>Mršić,	Leo	(Avtor)
	</dc:creator><dc:creator>Kalampakas,	Antonios	(Avtor)
	</dc:creator><dc:creator>Allahviranloo,	Tofigh	(Avtor)
	</dc:creator><dc:subject>link prediction</dc:subject><dc:subject>similarity indices</dc:subject><dc:subject>fuzzy social network</dc:subject><dc:subject>strength prominence (SP) index</dc:subject><dc:description>Link prediction is a field within social network studies that aims to forecast future connections based on the structure of a social network. This paper introduces a link prediction method based on the strength and prominence of seed node pairs, referred to as the strength prominence index. In this method, we get a consistent score for any pair of nodes, regardless of whether they share a common neighbour. Several key characteristics have been identified. In our experiments, we used three well-known estimators to evaluate the accuracy of link prediction algorithms: precision, area under the precision-recall curve, and area under the receiver operating characteristic curve. A comparative study with existing methods is also presented, supported by relevant graphs and tables. Validation using Facebook data sets demonstrates the effectiveness of the proposed method.</dc:description><dc:publisher>Springer</dc:publisher><dc:date>2025</dc:date><dc:date>2025-06-12 04:06:00</dc:date><dc:type>Neznano</dc:type><dc:identifier>22641</dc:identifier><dc:identifier>UDK: 519.876:316.7</dc:identifier><dc:identifier>ISSN pri članku: 2198-6053</dc:identifier><dc:identifier>DOI: 10.1007/s40747-025-01925-6</dc:identifier><dc:identifier>COBISS_ID: 239073539</dc:identifier><dc:language>sl</dc:language><dc:rights>© The Author(s)</dc:rights></metadata>
