Munich Personal RePEc Archive

International Scientific Co-Publications in Europe

Leogrande, Angelo and Costantiello, Alberto and Laureti, Lucio and Matarrese, Marco Maria (2022): International Scientific Co-Publications in Europe.

[thumbnail of MPRA_paper_113162.pdf]

Download (1MB) | Preview


The determinants of “International Scientific Co-Publications” in Europe are analyzed in the following article. Data from the European Innovation Scoreboard-EIS of the European Union for 36 countries in the period 2010-2019 were used. The data were analyzed using Panel Data with Fixed Effects, Panel Data with Random Effects, Pooled OLS, WLS, Dynamic Panel. The results show that the variable “International Scientific Co-Publication” is negatively associated with “Employment Share Manufacturing”, “Intellectual Assets”, “Turnover Share Large Enterprises”, “Linkages” and positively associated with “SMEs innovating in house”, “Trademark Applications”, “Human Resources”, “Publicprivate co-publications”, “Attractive Research Systems”, “Government procurement of advanced technology products”, “Turnover Share SMEs”. Then, a clustering analysis is realized with the algorithm k-Means. The Silhouette Coefficient and the Elbow Method are confronted to optimize the k-Means algorithm. The results show that the Elbow method is more efficient than the Silhouette coefficient in identifying the optimal number of clusters corresponding to k = 4 with the k-Means algorithm. A network analysis was then carried out using the “Manhattan Distance”. The analysis shows the presence of 9 network structures of which 5 are complex i.e., with a number of linkages greater than 3, and 4 are basic i.e. consist of a single link between two countries. Furthermore, we confront eight different machine learning algorithms to predict the future level of “International Scientific Co-Publications”. We found that the best algorithm in performing prediction with original data is the Tree Ensemble Regression. The predicted value of “International Scientific Co-Publication” is expected to growth by 0.61%.

Atom RSS 1.0 RSS 2.0

Contact us: mpra@ub.uni-muenchen.de

This repository has been built using EPrints software.

MPRA is a RePEc service hosted by Logo of the University Library LMU Munich.