Answers are grounded in this paper and cite their sources.
Virtual screening: Based on parameters which do not have to be measured, sets of attributes for solids are derived and used to predict whether a catalyst falls into one of five performance classes in propene oxidation, with a predictive power substantially exceeding the statistically expected values (see table (a confusion matrix): prediction rate indicates the correct assignment by the new method, ratio indicates the statistical expectation value).
No takes yet. Share an insight, caveat, or question.
Klanner et al. (2004) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: