[jira] Commented: (MAHOUT-146) Make Wikipedia Example Classifier more generic

["Robin Anil (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MAHOUT-146?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12732435#action_12732435 ] 

Robin Anil commented on MAHOUT-146:
-----------------------------------

Its already generic to some extend. Check out Mahout-60 

The usage was to create a dataset for BayesClassifier like this.
{noformat}
hadoop jar build/apache-mahout-0.1-dev-ex.jar org.apache.mahout.examples.classifiers.cbayes.WikipediaDatasetCreator -i wikipediadump -o wikipediainput -c pathto/country.txt
{noformat}

-c is the file with the list of categories(wikipedia categories). So you could specify any thing there. But it has to be a wikipedia category. When i go thought the xml dump, in the Map stage, for every article I match the list of categories with categories the document is in and output it if a match occurs

for example, you could create 2 categories by adding the following in the categories file
Scientists of 1900
Scientists of 2000

> Make Wikipedia Example Classifier more generic
> ----------------------------------------------
>
>                 Key: MAHOUT-146
>                 URL: https://issues.apache.org/jira/browse/MAHOUT-146
>             Project: Mahout
>          Issue Type: Improvement
>          Components: Classification
>            Reporter: Grant Ingersoll
>            Assignee: Grant Ingersoll
>            Priority: Minor
>             Fix For: 0.2
>
>
> It would be nice if the Wikipedia classifier example was a bit more generic instead of taking just countries.  For example, one could classify based on other types of categories, such as things like "subjects", i.e. History, Math, Science or other things.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.