The two components (paracetamol and amantadine hydrochloride) were simultaneously determined in combined paracetamol and amantadine hydrochloride tablets and powder by using near-infrared (NIR) spectroscopy and artificial neural networks (ANNs). The ANN models of three pretreated spectra (first-derivative, second-derivative and standard normal variate (SNV), respectively) were established. The mathematical corrected models of tablets were compared with those of the powder. In the models, the concentrations of paracetamol and amantadine hydrochloride as the active components were determined simultaneously and compared with the results of their separate determination. The parameters that affected the network were studied and the concentrations of the test set samples were predicted. The degree of approximation, a new evaluation criterion of the network was employed to prove the accuracy of the predicted results.