The quantitative structure property relationship (QSPR) for thermal conductivity of liquid aliphatic alcohols was developed on the basis of 139 thermal conductivity data points of liquid aliphatic alcohols, which were divided into a 65-member training set, a 20-member validation set, and a 54-member prediction set. Four parameters (temperature-T, the intrinsic state pseudoconnectivity index-type ls-Psi_i_ls, the sixth eigenvalue from augmented edge adjacency matrix weighed by edge degree-Eig06_AEA(ed), and the global topological charge index-JGT) were screened to develop the model by using the stepwise regression and the best subset regression method. For the training set, validation set and prediction set, the square correlation coefficient (R-2) is 0.9769, 0.9726, and 0.9738, respectively. The mean relative deviation values of training set, validation set, and prediction set were 1.4%, 1.6%, and 1.6%. The QSPR model can provide not only basic data for the engineering application but also theoretical guidance for designing and seeking specific thermal conductivity materials.