Imagine that you have been given a dataset of 1,000 documents that have been classified as being about entertainment or education. There are 700 entertainment documents in the dataset and 300 education documents in the dataset. The tables below give the number of documents from each topic that a selection of words occurred in. Word-document counts for the entertainment dataset fun christmas is 695 machine 35 family 400 learning 70 415 Word-document counts for the education dataset is christmas fun 200 machine 120 family 10 learning 105 295a. What target level will a naive Bayes model predict for the following query document: "machine learning is fun"? b. What target level will a naive Bayes model predict for the following query document: "christmas family fun"? c. What target level will a naive Bayes model predict for the query document in part (b) of this question, if Laplace smoothing with k = 10 and a vocabulary size of 6 is used?
