Transition Probability Matrix calculation for sentences

Question

I have sentences stored as strings extracted from a document. I want to apply standard cosine similarity to sentences. How do i go about doing it?

Answer 1

Well you already have the formula on wikipedia. Each of the Ai is a word so first you need to compute the frequency of words in your document . You need to be able to create a map of word occurrences. Then you represent each sentences by a vector of words and you can apply the formula.

Answer 2

First of all, read about Term-Document matrix

Then, go for the Cosine calculation using Cosine Similarity calculator

If you are interested in text mining, then go for SVD and finally Latent Semantic Analysis