mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Suneel Marthi <suneel_mar...@yahoo.com>
Subject Re: Question re cluster-reuters.sh
Date Tue, 10 Sep 2013 14:26:24 GMT
Michael.

1. build-reuters.sh is to be be retired, use cluster-reuters.sh instead.
2. You are correct, the script does what's been described in the wiki link.


________________________________
 From: Michael Wechner <michael.wechner@wyona.com>
To: mahout-user@apache.org 
Sent: Tuesday, September 10, 2013 9:55 AM
Subject: Question re cluster-reuters.sh
 

Hi

I have tried to follow/execute the steps described at

https://cwiki.apache.org/confluence/display/MAHOUT/Quick+tour+of+text+analysis+using+the+Mahout+command+line

but had trouble to do so, because for example |org.apache.lucene.benchmark.utils.ExtractReuters
does not seem to beĀ  contained by mahout-distribution-0.8.

Then I have found

mahout-distribution-0.8/examples/bin/build-reuters.sh

and was running it successfully.

If I understand the script is doing basically the same as described at

|https://cwiki.apache.org/confluence/display/MAHOUT/Quick+tour+of+text+analysis+using+the+Mahout+command+line

right? If so, then I would be happy to update the Wiki accordingly.

Also I have found the following post

http://bickson.blogspot.ch/2011/09/understanding-mahout-k-means-clustering.html

which seems to describe quite nicely what is happening behind the scenes.

I think it would also make sense to add these notes to the Wiki, or WDYT?

Thanks

Michael
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message