An Attempt to Boost Molecular Descriptors with Quantum-Derived Features in Prediction of Maximum Emission Wavelengths of Chromophores

Abstract

The following research assesses the capability of machine learning in predicting maximum emission wavelength of organic compounds. The predictions are based on structure descriptors and fingerprints widely applied in cheminformatics. In an attempt to further improve accuracy, developed machine learning models were enriched with quantum mechanics derived features. Multi linear, gradient boosting and random forest regressions were applied. Computers were trained and tested with database of experimental data of optical properties.

Toward Quantum-Informed Atom Pairs

“…25 The code in the form of Jupyter Notebooks and Python scripts is available at GitHub. 35 Data sets limited to the applicability domain are attached to this article.…”