Consider the below input and answer the following questions. [1+1+2 = 4 Marks] Document 1 (d1) Document 2 (d2) Document 3 (d3) Document 4 (d4) Jack bought jacket Jacket had hood Jill wore jacket Jill hated hood a) List only one sample feature vector for each of the below use cases using the above training data: 1. Case 1: Requirement is to automate document clustering 2. Case 2: Requirement is to model a system that can predict & recommend synonym of words b) Is document d4 similar to d1 or d2? Justify your answer using cosine similarity measure. Note: Ignore punctuations & treat small case vs capital case as same but retain all the tokens, including stop words if any.