Modeling from theory and modeling from data: complementary or alternative approaches? The case of ionic liquids

In the field of ionic liquids (ILs) theory-driven modelling approaches aimed at the best fit of all available data by a unique often non linear model have been widely adopted to develop Quantitative Structure-Property Relationships (QSPR) models. In this context, we propose cheminformatics and chemometrics data–driven procedures leading to QSPR soft models of local validity able to predict relevant ILs physico-chemical properties such as viscosity, density, decomposition temperature and conductivity. These models, using readily available and easily interpretable VolSurf+ descriptors, represent an unexploited opportunity for experimentalists to model and predict physico-chemical properties of ILs in industrial R&D design.