Comments on: Atbrox Customer Case Study – Scalable Language Processing with Elastic Mapreduce (Hadoop) http://atbrox.com/2009/11/14/atbrox-customer-case-study-scalable-language-processing-with-elastic-mapreduce-hadoop/ Wed, 25 Jan 2012 17:02:46 +0000 hourly 1 http://wordpress.org/?v=3.2.1 By: amund http://atbrox.com/2009/11/14/atbrox-customer-case-study-scalable-language-processing-with-elastic-mapreduce-hadoop/comment-page-1/#comment-184 amund Mon, 23 Nov 2009 20:38:35 +0000 http://atbrox.com/?p=507#comment-184 the scope of this project was the creation of n-gram and not the processing afterwards (except some filtering). the scope of this project was the creation of n-gram and not the processing afterwards (except some filtering).

]]> By: Øystein Nytrø http://atbrox.com/2009/11/14/atbrox-customer-case-study-scalable-language-processing-with-elastic-mapreduce-hadoop/comment-page-1/#comment-104 Øystein Nytrø Mon, 16 Nov 2009 20:15:21 +0000 http://atbrox.com/?p=507#comment-104 Referring to the customer case, I would guess that n-gram postprocessing and subsequent model-building are the most complex, and depending on n, computationally intensive tasks. Was that done locally afterwards, or did you manage to distribute that also? --- Øystein N. Referring to the customer case, I would guess that n-gram postprocessing and subsequent model-building are the most complex, and depending on n, computationally intensive tasks. Was that
done locally afterwards, or did you manage to distribute that also?

— Øystein N.

]]>