ankusa
ankusa copied to clipboard
bmuller
Text classifier in Ruby that uses Hadoop/HBase, Mongo, or Cassandra for storage. New location for http://github.com/livingsocial/ankusa
On Ruby 2.4.0, the `require 'ankusa'` line fails unless preceded by `require 'set'`
I'm not entirely sure what the cause could be here, but when running Ankusa from rubygems, any attempt to connect to Mongo returns this error `/Library/Ruby/Gems/2.0.0/gems/ankusa-0.1.0/lib/ankusa/mongo_db_storage.rb:12:in `initialize': uninitialized constant Mongo::Connection...
When I train the data and classify items, they work. When I store the classification(to file) and load like this: storage = Ankusa::FileSystemStorage.new 'c:\type_class_classification' c = Ankusa::NaiveBayesClassifier.new storage and try...
TextHash currently assumes all input is in english and will therefor atomize, stem and skip stopwords. As I'm using Ankusa for another language I'd prefer to skip these methods. They...
