mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Thomas <>
Subject How to create Vectors from a tokenStream
Date Wed, 10 Oct 2012 16:10:21 GMT

I have built a Naive Bayes model from my Solr's index 
and I am now trying to classify documents 
as they arrive in Solr using an UpdateRequestProcessorFactory.

I was able to retrieve the input field's content and get it through 
some Filters and tokenizers.

I have now a tokenStream and I think I should convert it to a Vector 
so I can use my NaiveBayesClassifier's method classifyFull(Vector instance).

But I can't figure out how to build this Vector !?

I would like to avoid writing seqfiles if possible.

View raw message