The training sets were prepared by spiking pork to beef meatball in concentration range of 10.0–90.0 %. Meatball containing of 100 % beef and 100 % pork was also made to observe the spectral differentiation. In order to reduce the effect of other composition that usually added during meatball preparation (e.g. starch, salt and certain spices) towards NIR spectra obtained. The other composition and beef content were varying, and used for classification as pure. Similarly, this was also applied for the pork adulterated meatball, and used for classification as adulterated. This procedure is needed to make sure that the different in NIR spectra is due to the different in beef and pork content in meatball and not other compositional changes. For test set, another series of meatball containing the mixture of pork and beef were prepared (20, 50 & 70 %). Meatball was further subjected to powder. The powders obtained were analyzed using NIR spectroscopy. The spectral regions where the variations were observed were chosen for developing PLS and classification model (i.e. linear discriminant analysis (LDA), soft independent modelling of class analogies (SIMCA) and support vector machine (SVM).