TY - JOUR

T1 - Meta-learning for estimating the energy gaps of aromatic molecules

AU - Petrosyan, L.S.

AU - Koskin, I.P.

AU - Kazantsev, M.S.

N1 - Meta-learning for estimating the energy gaps of aromatic molecules / L. S. Petrosyan, I. P. Koskin, M. S. Kazantsev // Chemistry for Sustainable Development. – 2025. – Vol. 33. - No. 5. – P. 555-561. – DOI 10.15372/CSD2025688. – EDN CHQTQL. The work was carried out within the State Assignment for the Novosibirsk Institute of Organic Chemistry SB RAS (Project No. FWUE-2025-0007).

PY - 2025

Y1 - 2025

N2 - A novel predictive model based on meta-learning algorithms for estimating the energy gap between the frontier molecular orbitals of aromatic π-conjugated molecules is presented. The main goal of the study was to develop a highly accurate and robust model capable of replacing computationally expensive quantum chemistry calculations, in particular the methods based on density functional theory, for screening organic compounds in optoelectronic applications. The filtered subset of the publicly available PubChemQC PM6 database was used as the primary dataset. Molecular structure was encoded using Morgan fingerprints, which served as input for training three base models: Random Forest, Gradient Boosted Trees, and a fully connected Neural Network. Among these, the Gradient Boosted Trees model achieved the best performance (the mean absolute error was 0.1795 eV). To improve prediction accuracy, a meta-model was implemented and trained on the outputs of the above-mentioned base models. This approach demonstrated improved accuracy: the final mean absolute error was 0.1744 eV, which is 8 % better than simple averaging and 3 % better than the best-performing individual model. The proposed approach can be further enhanced by expanding the dataset and incorporating additional models, which pave the way for more accurate and efficient prediction of the properties of organic conjugated molecules in optoelectronics.

AB - A novel predictive model based on meta-learning algorithms for estimating the energy gap between the frontier molecular orbitals of aromatic π-conjugated molecules is presented. The main goal of the study was to develop a highly accurate and robust model capable of replacing computationally expensive quantum chemistry calculations, in particular the methods based on density functional theory, for screening organic compounds in optoelectronic applications. The filtered subset of the publicly available PubChemQC PM6 database was used as the primary dataset. Molecular structure was encoded using Morgan fingerprints, which served as input for training three base models: Random Forest, Gradient Boosted Trees, and a fully connected Neural Network. Among these, the Gradient Boosted Trees model achieved the best performance (the mean absolute error was 0.1795 eV). To improve prediction accuracy, a meta-model was implemented and trained on the outputs of the above-mentioned base models. This approach demonstrated improved accuracy: the final mean absolute error was 0.1744 eV, which is 8 % better than simple averaging and 3 % better than the best-performing individual model. The proposed approach can be further enhanced by expanding the dataset and incorporating additional models, which pave the way for more accurate and efficient prediction of the properties of organic conjugated molecules in optoelectronics.

KW - МАШИННОЕ ОБУЧЕНИЕ

KW - ЭНЕРГЕТИЧЕСКИЙ ЗАЗОР

KW - СОПРЯЖЕННЫЕ МОЛЕКУЛЫ

KW - МЕТА-ОБУЧЕНИЕ

KW - MACHINE LEARNING

KW - ENERGY GAP

KW - CONJUGATED MOLECULES

KW - META-LEARNING

UR - https://www.webofscience.com/wos/woscc/full-record/WOS:001677414900001

UR - https://www.elibrary.ru/item.asp?id=88834908

UR - https://www.mendeley.com/catalogue/ecbead04-3b5e-3e2b-991a-f8d29c126e51/

U2 - 10.15372/csd2025688

DO - 10.15372/csd2025688

M3 - Article

VL - 33

SP - 555

EP - 561

JO - Chemistry for Sustainable Development

JF - Chemistry for Sustainable Development

SN - 1817-1818

IS - 5

M1 - 8

ER -