Further improvements on estimating the popularity of recently published papers

• Published in: Quantitative science studies Vol. 2; no. 4; pp. 1529 - 1550
• Main Authors: Chatzopoulos, Serafeim, Vergoulis, Thanasis, Kanellos, Ilias, Dalamagas, Theodore, Tryfonopoulos, Christos
• Format: Journal Article
• Language: English
• Published: One Rogers Street, Cambridge, MA 02142-1209, USA MIT Press 04-02-2022; MIT Press Journals, The
• Subjects: Bibliometrics; Configurations; Estimation; Graphs; Knowledge representation; scholarly knowledge graphs; scientific impact assessment
• ISSN: 2641-3337; 2641-3337
• DOI: 10.1162/qss_a_00165

As the number of published scientific papers continually increases, the ability to assess their impact becomes more valuable than ever. In this work, we focus on the problem of estimating the expected citation-based popularity (or short-term impact) of papers. State-of-the-art methods for this problem attempt to leverage the current citation data of each paper. However, these methods are prone to inaccuracies for recently published papers, which have a limited citation history. In this context, we previously introduced
, an approach that can be applied on top of any popularity estimation method to improve its accuracy. Its power originates from providing more accurate estimations for the most recently published papers by considering the popularity of similar, older ones. In this work, we present
, an improved
adaptation that considers an additional type of paper similarity and
incorporates a faster configuration procedure, resulting in improved effectiveness and configuration efficiency.