Skip to main content Skip to main navigation


Estimating the Importance of Relational Features by Using Gradient Boosting

Matej Petkovic; Michelangelo Ceci; Kristian Kersting; Saso Dzeroski
In: Denis Helic; Gerhard Leitner; Martin Stettinger; Alexander Felfernig; Zbigniew W. Ras (Hrsg.). Foundations of Intelligent Systems - 25th International Symposium, ISMIS 2020, Proceedings. International Symposium on Methodologies for Intelligent Systems (ISMIS-2020), September 23-25, Graz, Austria, Pages 362-371, Lecture Notes in Computer Science (LNCS), Vol. 12117, Springer, 2020.


With data becoming more and more complex, the standard tabular data format often does not suffice to represent datasets. Richer representations, such as relational ones, are needed. However, a relational representation opens a much larger space of possible descriptors (features) of the examples that are to be classified. Consequently, it is important to assess which features are relevant (and to what extent) for predicting the target. In this work, we propose a novel relational feature ranking method that is based on our novel version of gradient-boosted relational trees and extends the Genie3 score towards relational data. By running the algorithm on six well-known benchmark problems, we show that it yields meaningful feature rankings, provided that the underlying classifier can learn the target concept successfully.

Weitere Links