Re: UIMA

2014-01-15 Thread Jens Grivolla
Hello Burcu, UIMA has an entirely different purpose actually, and doesn't do classification or clustering. You would rather use UIMA to enrich documents (individually) through text analysis and then use the result to create better feature vectors to use with Solr, Mahout, etc. We typically

Simple Recommendation System with Mahout and Netbeans

2014-01-15 Thread Chameera Wijebandara
Hi, I have just published the simple blog post which describe simple recommendation system with Mahout. Can you please tell me are there any thing to improve in thishttp://chameerawijebandara.wordpress.com/2014/01/15/simple-recommendation-system-with-mahout-and-netbeans/ blog post.

Re: Mahout 0.9 Release Candidate - VOTE

2014-01-15 Thread Chameera Wijebandara
Hi Tharindu, Still I could not able to download the artifacts. Could you please hep me to test the Release Thanks Chameera On Wed, Jan 15, 2014 at 12:21 PM, Suneel Marthi suneel_mar...@yahoo.comwrote: Thanks Tharindu. On Tuesday, January 14, 2014 11:30 PM, Tharindu Rusira

Re: UIMA

2014-01-15 Thread Burcu B
Hi, Thank you, Jens. I was planning to use OpenNLP for named entity recognition directly for the analysis you''ve mentioned; and Lucene for tokenization. However, UIMA has OpenNLP component, too. What is the reason to use UIMA instead of uisng OpenNLP and SOLR together? I am planning to use

Re: Simple Recommendation System with Mahout and Netbeans

2014-01-15 Thread Manuel Blechschmidt
Hi Chameera, On 15.01.2014, at 13:59, Chameera Wijebandara wrote: Hi, I have just published the simple blog post which describe simple recommendation system with Mahout. Can you please tell me are there any thing to improve in

Re: UIMA

2014-01-15 Thread Jens Grivolla
Hi, the advantage of using UIMA over plain OpenNLP is that it can allow you to more easily combine components from different sources, e.g. a tokenizer and POS tagger from OpenNLP, a parser from Stanford, etc. You then have components for input that deal with the different sources, and for

Re: Simple Recommendation System with Mahout and Netbeans

2014-01-15 Thread Chameera Wijebandara
Hi Manuel, Sry I didn't get your comment. Could you please explain father more? Thanks, Chameera On Wed, Jan 15, 2014 at 7:24 PM, Manuel Blechschmidt manuel.blechschm...@gmx.de wrote: Hi Chameera, On 15.01.2014, at 13:59, Chameera Wijebandara wrote: Hi, I have just published

Re: Simple Recommendation System with Mahout and Netbeans

2014-01-15 Thread Manuel Blechschmidt
Hi Chameera, Netbeans has excellent maven support http://wiki.netbeans.org/Maven Mahout uses maven to build their artifacts a.k.a. jars. They get deployed to the central repositories. There is no need to create custom builds. On the following screenshot you already see the wizard for a maven

Re: Simple Recommendation System with Mahout and Netbeans

2014-01-15 Thread Chameera Wijebandara
Hi Manuel, Thank you very mutch. I got it I'll corect it Thanks, Chameera On Wed, Jan 15, 2014 at 7:47 PM, Manuel Blechschmidt manuel.blechschm...@gmx.de wrote: Hi Chameera, Netbeans has excellent maven support http://wiki.netbeans.org/Maven Mahout uses maven to build their artifacts

Re: Simple Recommendation System with Mahout and Netbeans

2014-01-15 Thread Chameera Wijebandara
Hi Manuel, Thank you very mutch. I got it I'll corect it Thanks, Chameera On Wed, Jan 15, 2014 at 7:47 PM, Manuel Blechschmidt manuel.blechschm...@gmx.de wrote: Hi Chameera, Netbeans has excellent maven support http://wiki.netbeans.org/Maven Mahout uses maven to build their artifacts

Re: Item recommendation w/o users or preferences

2014-01-15 Thread Pat Ferrel
Haven’t read the whole thread but it sounds like you just need some simple start-here info… To do collaborative filtering you must have user-id, item-id, action/weight. For a minimal commerce CF cooccurrence recommender this is typically something like: user-id, item-id, 1=purchase To use

seqdirectory error on mapreduce

2014-01-15 Thread pzhang
My set-ups seems work well: $hadoop fs –copyFromLocal reuters_text reuters_text $hadoop fs –ls reuters_text/ Many files in the directory $mahout seqdirectory –i reuters_text –o reuters_seq –c UTF-8 reuters_seq directory generated but empty

Re: Mahout 0.9 Release Candidate - VOTE

2014-01-15 Thread Suneel Marthi
The Release has been rolled back to include a few last minute fixes, will be sending out a new Release link in a day. On Wednesday, January 15, 2014 1:33 PM, spa...@gmail.com spa...@gmail.com wrote: I am getting a 404 Item not found exception for this link. Complete stack trace is - 404 -

Classification of books

2014-01-15 Thread Suresh M
Hi, Our application will be getting books from different users. We have to classify them accordingly. Some one please tell me how to do that using apache mahout and java. Is hadoop necessary for that? -- Thank Regards Suresh

Re: Classification of books

2014-01-15 Thread Suresh M
Hi, Can you please tell me what does that pre-processing mean? Is it vectorization(as explained in Mahout in Action book) Can it be done using java and Mahout AP ? And, the model means, is it a class? On 16 January 2014 11:38, KK R kirubakumar...@gmail.com wrote: Hi Suresh, Apache Mahout