Predicting Scientific Breakthroughs Based on Structural Dynamic of Citation Cascades

Houqiang Yu, Yian Liang and Yinghua Xie ()
Additional contact information
Houqiang Yu: School of Information Management, Sun Yat-sen University, Guangzhou 510006, China
Yian Liang: School of Information Management, Sun Yat-sen University, Guangzhou 510006, China
Yinghua Xie: School of Information Management, Sun Yat-sen University, Guangzhou 510006, China

Mathematics, 2024, vol. 12, issue 11, 1-18

Abstract: Predicting breakthrough papers holds great significance; however, prior studies encountered challenges in this task, indicating a need for substantial improvement. We propose that the failure to capture the dynamic structural-evolutionary features of citation networks is one of the major reasons. To overcome this limitation, this paper introduces a new method for constructing citation cascades of focus papers, allowing the creation of a time-series-like set of citation cascades. Then, through a thorough review, three types of structural indicators in these citation networks that could reflect breakthroughs are identified, including certain basic topological metrics, PageRank values, and the von Neumann graph entropy. Based on the time-series-like set of citation cascades, the dynamic trajectories of these indicators are calculated and employed as predictors. Using the Nobel Prize-winning papers as a landmark dataset, our prediction method yields approximately a 7% improvement in the ROC-AUC score compared to static-based prior methods. Additionally, our method advances in achieving earlier predictions than other previous methods. The main contribution of this paper is proposing a novel method for creating citation cascades in chronological order and confirming the significance of predicting breakthroughs from a dynamic structural perspective.

Keywords: predictions; breakthroughs; networks; structure; dynamics (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/12/11/1741/pdf (application/pdf)
https://www.mdpi.com/2227-7390/12/11/1741/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:12:y:2024:i:11:p:1741-:d:1407906

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().