POS Tagging Using Hidden Markov Models in Natural Language Processing

In this paper, we thoroughly examine the fundamental part which is Parts-Of-Speech (POS) Tagging of Natural Language Processing (NLP), especially. Our examination revolves around Hidden Markov Models (HMMs) usage to face up to this crucial element of language comprehension. Starting with the fundamental principles of the NLP, we go through its history after that, and we explore the inner workings of the N-gram Models and the Markov Models with a more detailed view finally in Hidden Markov Models. We meticulously uncover approaches used in POS Tagging preparation, so that we can get down to the in-depth methodological part of the journey. The intricate workings of Viterbi Algorithm and Maximum Entropy Markov Models will be clarified in the methodology section as they are keystone in our POS Tagging exploits. By performing highly standardized experimentation, we check the competence of HMMs in POS tagging, separately with CRFs (Conditional Random Fields). Ultimately, the report is a comprehensive description of the use of Hidden Markov Models in POS Tagging, that presents the applications, theoretical and practical bases, and the implications in the area of natural language understanding.