Show HN: Naive Bayes classifier for text categorization in five steps
towardsdatascience.com
Show HN: Naive Bayes classifier for text categorization in five steps
1–7 of 7 posts
Re: Show HN: Naive Bayes classifier for text categorization in five steps
#2From the article:
For an English spam classifier that
considers all the words in the English
language, the number of the words (n)
is approximately 171,476.
Re: Show HN: Naive Bayes classifier for text categorization in five steps
#3This is not a bad explanation but when doing this practically it can be useful to take log() of the probabilities so that you work with sums of logs rather than multiplying small floats.
Re: Show HN: Naive Bayes classifier for text categorization in five steps
#4This is not a bad explanation but when doing this practically it can be useful to take log() of the probabilities so that you work with sums of logs rather than multiplying small floats. http://getpopfile.org/docs/faq:bayesandlogs
Thank you for the insight John, I have included your remark on the article.
Re: Show HN: Naive Bayes classifier for text categorization in five steps
#5From the article: For an English spam classifier that considers all the words in the English language, the number of the words (n) is approximately 171,476. That's a remarkably precise number to be preceded by the word "approximately".
Agree, that was odd, it now reads:
"the number of the words (n) is approximately 170k"
Thank you for the remark.
Re: Show HN: Naive Bayes classifier for text categorization in five steps
#6Nice article, very glad to read it. Keep up the good work.
Re: Show HN: Naive Bayes classifier for text categorization in five steps
#7Nice article, very glad to read it. Keep up the good work.
Thank you Victor, I enjoyed your JS articles too!