Comparative Analyses of Data Driven Machine Learning Models for TADF Emitters

Thermally activated delayed fluorescence (TADF) chromophores have attracted significant attention because they can harvest singlets and triplets in organic light-emitting diodes (OLEDs), resulting in high external quantum efficiency (EQE). This work aims to use a data-driven machine-learning model to predict the relationship between EQE and essential features of TADF-based OLEDs. The study uses a set of experimental data and applies various machine-learning models to analyze the relationship between EQE and the features of the device. The Random Forest model is found to have the highest accuracy for predicting EQE for non-doped TADF emitters. The importance of feature selection and its correlation with EQE is also analyzed, providing insight into how to select the best machine-learning model for rapid material screening and device optimization for non-doped TADF materials.