Jia-Nan Wang, Jun-Ling Jin, Yun Geng, Shi-Ling Sun, Hong-Liang Xu, Ying-Hua Lu and Zhong-Min Su

The electronic excitation energies of 90 BODIPY derivatives were calculated by extreme learning machine neural network, which is more efficient and accurate than methods such as B3LYP, NN, and GANN. Four groups of descriptors were considered when building the predication model, and results showed that quantum chemical descriptions play most important role in predicting. A user-friendly web server, EEEBPre, was built for prediction, freely accessible to public.