fselector 1.3.1 → 1.4.0

data/ChangeLog

@@ -1,3 +1,9 @@
+2012-11-05	v1.4.0
+------------------
+
+  * update according to reviewers' comments
+  * several bug fixes based on users' feedback
+  
 2012-05-31	v1.3.1
 ------------------
 

data/README.md

@@ -1,21 +1,22 @@
-FSelector: a Ruby gem for feature selection and ranking
-===========================================================
+FSelector: a Ruby gem for feature selection
+===========================================
 
 **Home**: [https://rubygems.org/gems/fselector](https://rubygems.org/gems/fselector)  
 **Source Code**: [https://github.com/need47/fselector](https://github.com/need47/fselector)  
-**Documentation** [http://rubydoc.info/gems/fselector/frames](http://rubydoc.info/gems/fselector/frames)  
+**Documentation**: [http://rubydoc.info/gems/fselector/frames](http://rubydoc.info/gems/fselector/frames)  
+**Publication**: [Bioinformatics, 2012, 28, 2851-2852](http://bioinformatics.oxfordjournals.org/content/28/21/2851)  
 **Author**: Tiejun Cheng  
 **Email**: [need47@gmail.com](mailto:need47@gmail.com)  
 **Copyright**: 2012  
 **License**: MIT License  
-**Latest Version**: 1.3.1  
-**Release Date**: 2012-05-31
+**Latest Version**: 1.4.0  
+**Release Date**: 2012-11-05
 
 Synopsis
 --------
 
 FSelector is a Ruby gem that aims to integrate various feature 
-selection/ranking algorithms and related functions into one single 
+selection algorithms and related functions into one single 
 package. Welcome to contact me (need47@gmail.com) if you'd like to 
 contribute your own algorithms or report a bug. FSelector allows user 
 to perform feature selection by using either a single algorithm or an 
@@ -49,18 +50,18 @@ Feature List
     Accuracy                         Acc         weighting  multi-class    discrete
     AccuracyBalanced                 Acc2        weighting  multi-class    discrete
     BiNormalSeparation               BNS         weighting  multi-class    discrete
-    CFS_d                            CFS_d       subset     multi-class    discrete
+    CFS_d                            CFS_d       searching  multi-class    discrete
     ChiSquaredTest                   CHI         weighting  multi-class    discrete
     CorrelationCoefficient           CC          weighting  multi-class    discrete
     DocumentFrequency                DF          weighting  multi-class    discrete
     F1Measure                        F1          weighting  multi-class    discrete
     FishersExactTest                 FET         weighting  multi-class    discrete
-    FastCorrelationBasedFilter       FCBF        subset     multi-class    discrete
+    FastCorrelationBasedFilter       FCBF        searching  multi-class    discrete
     GiniIndex                        GI          weighting  multi-class    discrete
     GMean                            GM          weighting  multi-class    discrete
     GSSCoefficient                   GSS         weighting  multi-class    discrete
     InformationGain                  IG          weighting  multi-class    discrete
-    INTERACT                         INTERACT    subset     multi-class    discrete
+    INTERACT                         INTERACT    searching  multi-class    discrete
     JMeasure                         JM          weighting  multi-class    discrete
     KLDivergence                     KLD         weighting  multi-class    discrete
     MatthewsCorrelationCoefficient   MCC, PHI    weighting  multi-class    discrete
@@ -78,25 +79,25 @@ Feature List
     Specificity                      SP          weighting  multi-class    discrete
     SymmetricalUncertainty           SU          weighting  multi-class    discrete
     BetweenWithinClassesSumOfSquare  BSS_WSS     weighting  multi-class    continuous
-    CFS_c                            CFS_c       subset     multi-class    continuous
+    CFS_c                            CFS_c       searching  multi-class    continuous
     FTest                            FT          weighting  multi-class    continuous
-    KS_CCBF                          KS_CCBF     subset     multi-class    continuous
+    KS_CCBF                          KS_CCBF     searching  multi-class    continuous
     KSTest                           KST         weighting  two-class      continuous
     PMetric                          PM          weighting  two-class      continuous
     Relief_c                         Relief_c    weighting  two-class      continuous
     ReliefF_c                        ReliefF_c   weighting  multi-class    continuous
     TScore                           TS          weighting  two-class      continuous
     WilcoxonRankSum                  WRS         weighting  two-class      continuous
-    LasVegasFilter                   LVF         subset     multi-class    discrete, continuous, mixed
-    LasVegasIncremental              LVI         subset     multi-class    discrete, continuous, mixed
+    LasVegasFilter                   LVF         searching  multi-class    discrete, continuous, mixed
+    LasVegasIncremental              LVI         searching  multi-class    discrete, continuous, mixed
     Random                           Rand        weighting  multi-class    discrete, continuous, mixed
-    RandomSubset                     RandS       subset     multi-class    discrete, continuous, mixed
+    RandomSubset                     RandS       searching  multi-class    discrete, continuous, mixed
     
   **note for feature selection interface:**   
-  there are two types of filter methods, i.e., feature weighting algorithms and feature subset selection algorithms  
+  there are two types of filter algorithms: filter\_by\_feature\_weighting and filter\_by\_feature\_searching  
   
-  - for weighting type: use either **select\_feature\_by\_score!** or **select\_feature\_by\_rank!**  
-  - for subset type: use **select\_feature!**
+  - for former: use either **select\_feature\_by\_score!** or **select\_feature\_by\_rank!**  
+  - for latter: use **select\_feature!**
 
 **3. feature selection approaches**
 
@@ -205,7 +206,7 @@ Usage
     # an ensemble of 40 feature selectors with 90% data by random sampling
     re = FSelector::EnsembleSingle.new(r, 40, 0.90, :random_sampling)
 
-    # read SPECT data set  (under the test/ directory)
+    # read SPECT data set (under the test/ directory)
     re.data_from_csv('test/SPECT_train.csv')
 
     # number of features before feature selection
@@ -225,7 +226,7 @@ Usage
 	# creating an ensemble of feature selectors by using 
 	# two feature selection algorithms: InformationGain (IG) and Relief_d. 
 	# note: can be 2+ algorithms, as long as they are of the same type, 
-	# either feature weighting or feature subset selection algorithms
+	# either filter_by_feature_weighting or filter_by_feature_searching
 	
 	# test for the type of feature weighting algorithms 
     r1 = FSelector::IG.new

data/lib/fselector.rb

@@ -7,7 +7,7 @@ R.eval 'options(warn = -1)' # suppress R warnings
 #
 module FSelector
   # module version
-  VERSION = '1.3.1'
+  VERSION = '1.4.0'
 end
 
 # the root dir of FSelector

data/lib/fselector/algo_base/base.rb

@@ -343,7 +343,7 @@ module FSelector
     #   the subset selection type of algorithms, see {file:README.md}
     #
     def select_feature!
-      if not self.algo_type == :feature_subset_selection
+      if not self.algo_type == :filter_by_feature_searching
         abort "[#{__FILE__}@#{__LINE__}]: \n"+
               "  select_feature! is the interface for the type of feature subset selection algorithms only. \n" +
               "  please consider select_featue_by_score! or select_feature_by_rank!, \n" +
@@ -377,7 +377,7 @@ module FSelector
     #   the weighting type of algorithms, see {file:README.md}
     #
     def select_feature_by_score!(criterion, my_scores=nil)
-      if not self.algo_type == :feature_weighting
+      if not self.algo_type == :filter_by_feature_weighting
         abort "[#{__FILE__}@#{__LINE__}]: \n"+
               "  select_feature_by_score! is the interface for the type of feature weighting algorithms only. \n" +
               "  please consider select_featue!, \n" +
@@ -411,7 +411,7 @@ module FSelector
     #   the weighting type of algorithms, see {file:README.md}
     #
     def select_feature_by_rank!(criterion, my_ranks=nil)
-      if not self.algo_type == :feature_weighting
+      if not self.algo_type == :filter_by_feature_weighting
         abort "[#{__FILE__}@#{__LINE__}]: \n"+
               "  select_feature_by_rank! is the interface for the type of feature weighting algorithms only. \n" +
               "  please consider select_featue!, \n" +

data/lib/fselector/algo_both/LasVegasFilter.rb

@@ -15,7 +15,7 @@ module FSelector
     include Consistency
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_both/LasVegasIncremental.rb

@@ -13,7 +13,7 @@ module FSelector
     include Consistency
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_both/Random.rb

@@ -12,7 +12,7 @@ module FSelector
 #
   class Random < Base
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_both/RandomSubset.rb

@@ -10,7 +10,7 @@ module FSelector
 #
   class RandomSubset < Base
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_continuous/BSS_WSS.rb

@@ -19,7 +19,7 @@ module FSelector
 #
   class BetweenWithinClassesSumOfSquare < BaseContinuous
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private    
     

data/lib/fselector/algo_continuous/CFS_c.rb

@@ -13,7 +13,7 @@ module FSelector
     include Normalizer
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     private
     

data/lib/fselector/algo_continuous/F-Test.rb

@@ -24,7 +24,7 @@ module FSelector
 #
   class FTest < BaseContinuous
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
         
     private    
     

data/lib/fselector/algo_continuous/KS-CCBF.rb

@@ -13,7 +13,7 @@ module FSelector
     include Discretizer
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
         
     #
     # initialize from an existing data structure

data/lib/fselector/algo_continuous/KS-Test.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class KSTest < BaseContinuous
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
         
     private    
     

data/lib/fselector/algo_continuous/P-Metric.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class PMetric < BaseContinuous
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private    
     

data/lib/fselector/algo_continuous/ReliefF_c.rb

@@ -14,7 +14,7 @@ module FSelector
     include Normalizer
     
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private    
     

data/lib/fselector/algo_continuous/Relief_c.rb

@@ -14,7 +14,7 @@ module FSelector
     include Normalizer
     
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
        
     private
     

data/lib/fselector/algo_continuous/T-Score.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class TScore < BaseContinuous
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
      
     private    
     

data/lib/fselector/algo_continuous/WilcoxonRankSum.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class WilcoxonRankSum < BaseContinuous
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private    
     

data/lib/fselector/algo_discrete/Accuracy.rb

@@ -8,10 +8,12 @@ module FSelector
 #               tp+tn          A+D
 #     Acc = ------------- = ---------
 #            tp+fn+tn+fp     A+B+C+D
+#
+# ref: [An extensive empirical study of feature selection metrics for text classification](http://dl.acm.org/citation.cfm?id=944974)
 #
   class Accuracy < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/AccuracyBalanced.rb

@@ -11,7 +11,7 @@ module FSelector
 #
   class AccuracyBalanced < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/BiNormalSeparation.rb

@@ -14,7 +14,7 @@ module FSelector
 #
   class BiNormalSeparation < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
      
     private
     

data/lib/fselector/algo_discrete/CFS_d.rb

@@ -14,7 +14,7 @@ module FSelector
     include Entropy
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     private
     

data/lib/fselector/algo_discrete/ChiSquaredTest.rb

@@ -20,7 +20,7 @@ module FSelector
 #
   class ChiSquaredTest < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_discrete/CorrelationCoefficient.rb

@@ -14,7 +14,7 @@ module FSelector
 #
   class CorrelationCoefficient < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
         

data/lib/fselector/algo_discrete/DocumentFrequency.rb

@@ -11,7 +11,7 @@ module FSelector
 #
   class DocumentFrequency < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/F1Measure.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class F1Measure < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/FastCorrelationBasedFilter.rb

@@ -13,7 +13,7 @@ module FSelector
     include Entropy
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_discrete/FishersExactTest.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class FishersExactTest < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
        
     private
         

data/lib/fselector/algo_discrete/G-Mean.rb

@@ -13,7 +13,7 @@ module FSelector
 #
   class GMean < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/GSSCoefficient.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class GSSCoefficient < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/GiniIndex.rb

@@ -14,7 +14,7 @@ module FSelector
 #
   class GiniIndex < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/INTERACT.rb

@@ -14,7 +14,7 @@ module FSelector
     include Consistency
     
     # this algo outputs a subset of feature
-    @algo_type = :feature_subset_selection
+    @algo_type = :filter_by_feature_searching
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_discrete/InformationGain.rb

@@ -18,7 +18,7 @@ module FSelector
     include Entropy
 
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
   

data/lib/fselector/algo_discrete/J-Measure.rb

@@ -13,7 +13,7 @@ module FSelector
 #
   class JMeasure < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
         
     private    
     

data/lib/fselector/algo_discrete/KL-Divergence.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class KLDivergence < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
         
     private    
     

data/lib/fselector/algo_discrete/MatthewsCorrelationCoefficient.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class MatthewsCorrelationCoefficient < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/McNemarsTest.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class McNemarsTest < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     #
     # intialize from an existing data structure

data/lib/fselector/algo_discrete/MutualInformation.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class MutualInformation < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
        
     private
     

data/lib/fselector/algo_discrete/OddsRatio.rb

@@ -17,7 +17,7 @@ module FSelector
 #
   class OddsRatio < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/OddsRatioNumerator.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class OddsRatioNumerator < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/Power.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class Power < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     #
     # initialize from an existing data structure

data/lib/fselector/algo_discrete/Precision.rb

@@ -11,7 +11,7 @@ module FSelector
 #
   class Precision < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/ProbabilityRatio.rb

@@ -15,7 +15,7 @@ module FSelector
 #
   class ProbabilityRatio < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/ReliefF_d.rb

@@ -13,7 +13,7 @@ module FSelector
     include Discretizer
     
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
        
     private
     

data/lib/fselector/algo_discrete/Relief_d.rb

@@ -14,7 +14,7 @@ module FSelector
     include Discretizer
     
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private    
     

data/lib/fselector/algo_discrete/Sensitivity.rb

@@ -13,7 +13,7 @@ module FSelector
 #
   class Sensitivity < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/Specificity.rb

@@ -13,7 +13,7 @@ module FSelector
 #
   class Specificity < BaseDiscrete
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
     

data/lib/fselector/algo_discrete/SymmetricalUncertainty.rb

@@ -21,7 +21,7 @@ module FSelector
     include Entropy
 
     # this algo outputs weight for each feature
-    @algo_type = :feature_weighting
+    @algo_type = :filter_by_feature_weighting
     
     private
   

data/lib/fselector/fileio.rb

@@ -107,6 +107,11 @@ module FileIO
     ifs.close if not ifs == $stdin
     
     set_data(data)
+    
+    # feature name-type pairs
+    each_feature do |f|
+      set_feature_type(f, :numeric)
+    end
   end # data_from_libsvm
   
   
@@ -134,7 +139,13 @@ module FileIO
     each_sample do |k, s|
       ofs.print "#{k2idx[k]} "
       s.keys.sort { |x, y| f2idx[x] <=> f2idx[y] }.each do |f|
-        ofs.print " #{f2idx[f]}:#{s[f]}" if not s[f].zero? # implicit mode
+        if not s[f].is_a? Numeric
+          abort "[#{__FILE__}@#{__LINE__}]: \n"+
+                  "  LibSVM format only supports the following feature type: \n"+
+                  "  integer, real, numeric, float, double, continuous"
+        else
+          ofs.print " #{f2idx[f]}:#{s[f]}" if not s[f].zero? # implicit mode
+        end
       end
       ofs.puts
     end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: fselector
 version: !ruby/object:Gem::Version
-  version: 1.3.1
+  version: 1.4.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-05-31 00:00:00.000000000 Z
+date: 2012-11-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rinruby
-  requirement: &25065000 !ruby/object:Gem::Requirement
+  requirement: &24606276 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,8 +21,8 @@ dependencies:
         version: 2.0.2
   type: :runtime
   prerelease: false
-  version_requirements: *25065000
-description: FSelector is a Ruby gem that aims to integrate various feature selection/ranking
+  version_requirements: *24606276
+description: FSelector is a Ruby gem that aims to integrate various feature selection
   algorithms and related functions into one single package. Welcome to contact me
   (need47@gmail.com) if you'd like to contribute your own algorithms or report a bug.
   FSelector allows user to perform feature selection by using either a single algorithm
@@ -125,5 +125,5 @@ rubyforge_project:
 rubygems_version: 1.8.17
 signing_key: 
 specification_version: 3
-summary: feature selection and ranking
+summary: ! 'FSelector: a Ruby gem for feature selection'
 test_files: []