Machine learning to predict the partnership among universities

The 16th  RISIS Research Seminar will take place on 9th March from 12.30 to 2.00 pm (CET)  and will focus on a presentation entitled Link prediction in knowledge networks using exogenous and endogenous attributes: a machine learning approach, by Antonio Zinilli e Giovanni Cerulli, CNR-IRCrES.  Mike Thelwall, University of Wolverhampton  is involved as discussant.


The researchers propose a supervised machine learning approach to predict partnership formation between universities. The focus is  on successful joint R&D projects funded by Horizon 2020 programme in three research domains: Social Sciences and Humanities, Physical and Engineering Sciences, and Life Sciences.



The researchers perform two connected analyses: link formation prediction, and feature importance detection. As for link prediction, using out-of-sample cross-validated accuracy and a set of network endogenous and exogenous attributes, the researchers obtain 90% prediction accuracy when both types of attributes are used, and around 65% when using only the exogenous ones. This proves that partnership predictive power is on average 25% larger for universities already incumbent in the programme than for newcomers. As for feature importance, by computing super-learner average partial effects and elasticities, the study finds that the endogenous attributes are the most relevant in affecting the probability to generate a link and observe a largely negative elasticity of the link probability to feature changes, fairly uniform across attributes and domains.