Timeline for Parameters in Naive Bayes
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 23, 2021 at 7:53 | history | edited | Tim | CC BY-SA 4.0 |
added 1393 characters in body
|
| Aug 21, 2021 at 12:52 | comment | added | Arya McCarthy | Laplace smoothing for word counts may not be your best bet - see Whatβs wrong with adding one? (Gale and Church, 1994) | |
| Aug 21, 2021 at 10:15 | comment | added | Tim | @Esha this is maximum likelihood. In the documentation, they are saying that they use MAP so consider also priors. | |
| Aug 21, 2021 at 10:07 | comment | added | Esha | I am sorry, I don't quite understand... Why can't we estimate parameters through training data? Say we are building a email spam classifier using Naive Bayes. We have training data as 10000 emails. Now we pick all the spam mails (which is variable y) and calculate how many times a word (which is variable π₯π) is appearing in those mails. This would give us the probability π(π₯π|π¦). Where am I going wrong? | |
| Aug 21, 2021 at 9:34 | history | answered | Tim | CC BY-SA 4.0 |