The objective of this thesis is to redesign the relational structure of the IUPHAR database into a graph-based representation in order to overcome limitations in tracking and analyzing citation relationships. The relational nature of IUPHAR database, with its inherently hierarchical and tree-like organization, poses significant challenges for efficiently monitoring and collecting data on citations and references among database entries. To address this, the database is transformed into a graph model that enables the identification and extraction of both incoming and outgoing citations for all nodes, including references to external sources as well as citations within the database itself. Leveraging the concept of citation summaries, these relationships are systematically aggregated to construct graph objects corresponding to database nodes. Subsequently, comprehensive statistical analyses are performed to characterize the resulting citation network and to evaluate the influence of integrating citation summaries on author-level metrics, particularly the h-index, of researchers whose publications IUPHAR database entries cite. The findings offer insights into the structure and dynamics of citation relationships within IUPHAR database and demonstrate the potential of graph-based modeling for more effective bibliometric analysis in relationally organized scientific databases.
L’obiettivo di questa tesi è riprogettare la struttura relazionale del database IUPHAR in una rappresentazione basata su grafici, al fine di superare i limiti nel tracciamento e nell’analisi delle relazioni di citazione. La natura relazionale del database IUPHAR, con la sua organizzazione intrinsecamente gerarchica e ad albero, pone sfide significative per il monitoraggio e la raccolta efficiente dei dati sulle citazioni e i riferimenti tra le voci del database. Per ovviare a questo problema, il database viene trasformato in un modello grafico che consente l’identificazione e l’estrazione delle citazioni in entrata e in uscita per tutti i nodi, compresi i riferimenti a fonti esterne e le citazioni all’interno del database stesso. Sfruttando il concetto di sintesi delle citazioni, queste relazioni vengono aggregate sistematicamente per costruire oggetti grafici corrispondenti ai nodi del database. Successivamente, vengono eseguite analisi statistiche complete per caratterizzare la rete di citazioni risultante e valutare l’influenza dell’integrazione dei riassunti delle citazioni sulle metriche a livello di autore, in particolare l’indice h, dei ricercatori le cui pubblicazioni sono citate nelle voci del database IUPHAR. I risultati offrono approfondimenti sulla struttura e le dinamiche delle relazioni di citazione all’interno del database IUPHAR e dimostrano il potenziale della modellazione basata su grafici per un’analisi bibliometrica più efficace nei database scientifici organizzati in modo relazionale.
Graph-Based Representation and Exploration of Relational Data
TIKHONOV, VLADISLAV
2024/2025
Abstract
The objective of this thesis is to redesign the relational structure of the IUPHAR database into a graph-based representation in order to overcome limitations in tracking and analyzing citation relationships. The relational nature of IUPHAR database, with its inherently hierarchical and tree-like organization, poses significant challenges for efficiently monitoring and collecting data on citations and references among database entries. To address this, the database is transformed into a graph model that enables the identification and extraction of both incoming and outgoing citations for all nodes, including references to external sources as well as citations within the database itself. Leveraging the concept of citation summaries, these relationships are systematically aggregated to construct graph objects corresponding to database nodes. Subsequently, comprehensive statistical analyses are performed to characterize the resulting citation network and to evaluate the influence of integrating citation summaries on author-level metrics, particularly the h-index, of researchers whose publications IUPHAR database entries cite. The findings offer insights into the structure and dynamics of citation relationships within IUPHAR database and demonstrate the potential of graph-based modeling for more effective bibliometric analysis in relationally organized scientific databases.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tikhonov_Vladislav.pdf
Accesso riservato
Dimensione
1.6 MB
Formato
Adobe PDF
|
1.6 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/98454