Subject - Computer Science (Natural Language Processing) Early response appreciated, All answers will be upvoted.

a) Given a corpus C, the maximum likelihood estimation (MLE) for the bigram Hello World is 0.3 and the count of occurrence of the word Hello is 580 for the same corpus, the likelihood of Hello World" after applying the add-one smoothing is 0.04. What is the vocabulary size of Corpus C. [3 marks] b)What are the challenges in the Natural Language Processing? [3 marks] c)There were 100 documents and each document contained one word. 30 of these documents contained the word hello. I asked Bob to separate all the documents containing the word hello. He showed me 60 but hello" was not in 40 of them. Construct the confusion matrics and calculate the accuracy. [4 marks]