[
https://issues.apache.org/jira/browse/MAHOUT-1502?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13964963#comment-13964963
]
Andrew Palumbo commented on MAHOUT-1502:
----------------------------------------
Sounds good, Table 4 from the Rennie paper has a nice 8 step breakdown of the
algorithm. I would propose dropping table 4 in to take the place of the old
broken link (replacing TWCNB with CBayes to avoid confusion). It works out
nicely because steps 1-3 (1. TF transform, 2. IDF transform and 3. length
normalization) and are the now being done externally to NB and 4-8 are
internal. I will try to get this written up as quickly as possible. I'm pretty
well swamped from tomorrow afternoon on through the rest of the week. I hope
to get a draft out early next week.
> Update Naive Bayes Webpage to Current Implementation
> -----------------------------------------------------
>
> Key: MAHOUT-1502
> URL: https://issues.apache.org/jira/browse/MAHOUT-1502
> Project: Mahout
> Issue Type: Bug
> Components: Documentation
> Affects Versions: 0.9
> Reporter: Andrew Palumbo
> Priority: Minor
> Fix For: 1.0
>
>
> Current Naive Bayes page is for pre .7 NB implementation:
> https://mahout.apache.org/users/classification/bayesian.html
> post .7, TF-IDF calculations are preformed outside of NB.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
