create the TF-IDF matrix, creating training data set

1)create data set that has 10 emails, 5 about sad news and 5 about congratulatory notes. 2) for each email create a document vector containing not more than 12 important terms. 3) compute TF-IDF value for each term 4) Write a program that takes 12 terms as test input and then computes cosine similarity between the test input and each of the 10 document (email) vectors. If the document with the highest similarity is a “sad” email, the program outputs “sad”, otherwise the program outputs ”congratulatory”