(d) Describe how a Hidden Markov Model can be used to perform part-of- speech tagging based on a bigram model. Your answer should include details of how the model is trained using suitable data and explain how tagging can be achieved in time linearly related to sentence length. [10 marks]

(e) How would you go about providing a quantitative evaluation of the performance of a part-of-speech tagger? What is the likely impact of part-of-speech tagging errors on downstream processing steps? [10 marks]