1. Recommender Engine in Ruby. Based loosely in Apache Mahout
2. For using it, install the gem and then as an example do the following:

    def save_as_csv_file(file_path,values)
       File.open(file_path,'w') do |file|
         values.each do |row|
           file.puts "#{row[0]},#{row[1]},#{row[2]}"
         end
       end
    end    
    
    save_as_csv_file '/tmp/data_file',[['A','B',5],['A','C',3],['B','B',5],['B','C',3],['B','D',2]]
    data_model = Recommendations::DataModel::FileDataModel.new('/tmp/data_file')
    similarity = Recommendations::Similarity::EuclideanDistanceSimilarity.new(data_model)
    neighborhood = Recommendations::Similarity::Neighborhood::NearestNUserNeighborhood.new(data_model,similarity,5,0.5)
    rating_estimator = Recommendations::Recommender::Estimation::DefaultRatingEstimator.new(data_model,similarity)
    recommender = Recommendations::Recommender::GenericUserBasedRecommender.new(data_model,similarity,neighborhood,rating_estimator)
    recommendations = recommender.recommend('A',5)
    puts recommendations[0].item
    puts recommendations[0].value