Hidden Markov Models for POS tagging
Hidden Markov Models (HMM) are conducive to solving classification problems with generative sequences. In natural language processing, HMM can be used for a variety of tasks such as phrase chunking, parts of speech tagging, and information extraction from documents. If we consider words as input, while any prior information on the input can be considered as states, and estimated conditional probabilities can be considered as the output, then POS tagging can be categorized as a typical sequence classification problem that can be solved using HMM.
Basic definitions and notations
According to (Rabiner), there are five elements needed to define an HMM:
N denotes the number of states (which are hidden) in the model. For parts of speech tagging, N is the number of tags that can be used by the system. Each possible tag for the system corresponds to one state of the HMM. The possible interconnections of individual states are denoted by S = {S1,Sn}. Let qt denote...