In this era of machine learning and artificial intelligence, natural language processing applications are in high demand. For the development of almost all the natural language processing applications assignment of semantic informationis the fundamental component. This component assign the correct semantic information to individual word. In this research paper,we proposea stochastic approach for the development of system to assign correct semantic information for one of the morphologically rich Indian language i.e. Punjabi language. We have used annotated corpus available by Indian language corpora initiative to train the system. Thisdata is used for training and approximately 99k sentences collected form internet are used for testing.Further the developed system is also capable to assign the part of speech tag to unknown words. On testing the system, author claimed an accuracy of 89.33%.
Key words: Punjabi POS tagger, Stochastic, NLP, Grammar Checker