Benzenoid hydrocarbons are a group of the most important π-electron systems having the attention of both experimental and theoretical chemists for the last 100 years. In the present study, based on the general interaction properties function (GIPF) family descriptors, significant one- or two-parametric quantitative structure–property (activity) relationship models were developed for the prediction of properties/activities of benzenoids hydrocarbons. All descriptors were computed in density functional theory (DFT) at the B3LYP/STO-3G level of theory in Gaussian98 software. A large number of physico-chemical properties and two biological activities (e.g. bio-concentration factor and photo-induced toxicity) of these compounds were investigated by using multiple linear regressions. All created models were interpreted in term of selected descriptors. R2 and R2 cv values of all models are respectively between 0.665–0.994 and 0.609–0.990 for the whole dataset of each property/activity. Maximum R2 for Y-randomization (R2 max) test and its cross-validation (R2 cv,max) are between 0.098–0.485 and 0.002–0.357, respectively.
