spam-filtering - implementing Naive Bayes Theorem in Go

Heavily inspired by Tsoding's "Email Spam Filter in Go" video.

Formula:

$$ P(C|D) = \frac{P(D|C) \cdot P(C)}{P(D)} $$

where:

• $P(D)$ - probability of a given document to be exist based on the current Bag-of-Words (order of words is not relevant)
• $P(C)$ - probability of a class of document to be True (in this case the SPAM class)

get the dataset

Go to https://www2.aueb.gr/users/ion/data/enron-spam/ and download and extract enron1, enron2 etc. and put them in data/

Progress:

• generate Bag-of-Words for a directory
• compute $P(D)$ for a given document (at a filePath)
• compute $P(C)$
• compute $P(D|C)$