fselector 0.3.1 → 0.4.0

data/README.md

@@ -8,8 +8,8 @@ FSelector: a Ruby gem for feature selection and ranking
 **Email**: [need47@gmail.com](mailto:need47@gmail.com)  
 **Copyright**: 2012  
 **License**: MIT License  
-**Latest Version**: 0.3.1  
-**Release Date**: April 4 2012
+**Latest Version**: 0.4.0  
+**Release Date**: April 5 2012
 
 Synopsis
 --------
@@ -101,7 +101,7 @@ Feature List
     algorithm          note                                  feature type                     
     --------------------------------------------------------------------------------------
     fixed_value        replace with a fixed value            discrete, continuous
-    mean_value         replace with the mean feature value   continuous
+    mean_value         replace with mean feature value       continuous
     most_seen_value    replace with most seen feature value  discrete
 
 Installing
@@ -124,10 +124,10 @@ Usage
     # read from random data (or csv, libsvm, weka ARFF file)
     # no. of samples: 100
     # no. of classes: 2
-    # no. of features: 10
+    # no. of features: 15
     # no. of possible values for each feature: 3
     # allow missing values: true
-    r1.data_from_random(100, 2, 10, 3, true)
+    r1.data_from_random(100, 2, 15, 3, true)
     
     # number of features before feature selection
     puts "# features (before): "+ r1.get_features.size.to_s
@@ -141,7 +141,7 @@ Usage
     # you can also use multiple alogirithms in a tandem manner
     # e.g. use the ChiSquaredTest with Yates' continuity correction
     # initialize from r1's data
-    r2 = FSelector::ChiSquaredTest.new(:yates, r1.get_data)
+    r2 = FSelector::ChiSquaredTest.new(:yates_continuity_correction, r1.get_data)
     
     # number of features before feature selection
     puts "# features (before): "+ r2.get_features.size.to_s
@@ -157,7 +157,7 @@ Usage
     r2.data_to_weka(:stdout, :sparse)
 
 	
-**2. feature selection by an ensemble of algorithms**
+**2. feature selection by an ensemble of multiple algorithms**
 
     require 'fselector'
 	
@@ -169,7 +169,7 @@ Usage
     re = FSelector::Ensemble.new(r1, r2)
     
     # read random data
-    re.data_from_random(100, 2, 10, 3, true)
+    re.data_from_random(100, 2, 15, 3, true)
     
     # number of features before feature selection
     puts '# features (before): ' + re.get_features.size.to_s
@@ -185,7 +185,7 @@ Usage
     puts '# features (after): ' + re.get_features.size.to_s
 
     
- **3. normalization and discretization before feature selection**
+**3. normalization and discretization before feature selection**
 
  In addition to the algorithms designed for continuous feature, one
  can apply those deisgned for discrete feature after (optionally
@@ -194,14 +194,11 @@ Usage
     require 'fselector'
     
     # for continuous feature
-    r1 = FSelector::BaseContinuous.new
+    r1 = FSelector::Relief_c.new
     
     # read the Iris data set (under the test/ directory)
     r1.data_from_csv('test/iris.csv')
-    
-    # normalization by log2 (optional)
-    # r1.normalize_by_log!(2)
-    
+        
     # discretization by ChiMerge algorithm
     # chi-squared value = 4.60 for a three-class problem at alpha=0.10
     r1.discretize_by_ChiMerge!(4.60)
@@ -219,6 +216,8 @@ Usage
     # number of features after feature selection
     puts '# features (after): ' + r2.get_features.size.to_s
 
+**4. see more examples test_*.rb under the test/ directory**
+
 Copyright
 ---------
 FSelector © 2012 by [Tiejun Cheng](mailto:need47@gmail.com).

data/lib/fselector.rb

@@ -3,7 +3,7 @@
 #
 module FSelector
   # module version
-  VERSION = '0.3.1'
+  VERSION = '0.4.0'
 end
 
 ROOT = File.expand_path(File.dirname(__FILE__))
@@ -17,6 +17,10 @@ require "#{ROOT}/fselector/fileio.rb"
 require "#{ROOT}/fselector/util.rb"
 # entropy-related functions
 require "#{ROOT}/fselector/entropy.rb"
+# normalization for continuous data
+require "#{ROOT}/fselector/normalizer.rb"
+# discretization for continuous data
+require "#{ROOT}/fselector/discretizer.rb"
 # replace missing values
 require "#{ROOT}/fselector/replace_missing_values.rb"
 

data/lib/fselector/algo_base/base_continuous.rb

@@ -1,5 +1,3 @@
-require File.expand_path(File.dirname(__FILE__) + '/../algo_continuous/normalizer.rb')
-require File.expand_path(File.dirname(__FILE__) + '/../algo_continuous/discretizer.rb')
 #
 # FSelector: a Ruby gem for feature selection and ranking
 #

data/lib/fselector/algo_base/base_discrete.rb

@@ -41,10 +41,7 @@ module FSelector
     def get_A(f, k)
       @A ||= calc_A
       a = @A[k][f]
-      
-      # add 0.5 to avoid any ZERO in denominator or numerator
-      a+=0.5 if a.zero?
-      
+            
       a 
     end # get_A
 
@@ -78,9 +75,6 @@ module FSelector
     def get_B(f, k)
       @B ||= calc_B
       b = @B[k][f]
-          
-      # add 0.5 to avoid any ZERO in denominator or numerator
-      b+=0.5 if b.zero?
       
       b
     end # get_B
@@ -116,9 +110,6 @@ module FSelector
       @C ||= calc_C
       c = @C[k][f]
       
-      # add 0.5 to avoid any ZERO in denominator or numerator
-      c+=0.5 if c.zero?
-      
       c
     end # get_C
     
@@ -153,9 +144,6 @@ module FSelector
       @D ||= calc_D
       d = @D[k][f]
       
-      # add 0.5 to avoid any ZERO in denominator or numerator
-      d+=0.5 if d.zero?
-      
       d
     end # get_D
 

data/lib/fselector/algo_continuous/PMetric.rb

@@ -34,7 +34,11 @@ module FSelector
       end
       
       # calc
-      s = (s1.ave-s2.ave).abs / (s1.sd+s2.sd)
+      s = 0.0
+      dd = s1.sd+s2.sd
+      if not dd.zero?
+        s = (s1.ave-s2.ave).abs / dd
+      end
       
       set_feature_score(f, :BEST, s)      
     end # calc_contribution

data/lib/fselector/algo_continuous/TScore.rb

@@ -35,7 +35,14 @@ module FSelector
       
       # calc
       n1, n2 = s1.size, s2.size
-      s = (s1.ave-s2.ave).abs / Math.sqrt( (n1*s1.var+n2*s2.var) / (n1+n2) )
+      if not (n1+n2).zero?
+        dd = Math.sqrt( (n1*s1.var+n2*s2.var) / (n1+n2) )
+      end
+      
+      s = 0.0
+      if not dd.zero?
+        s = (s1.ave-s2.ave).abs / dd
+      end
       
       set_feature_score(f, :BEST, s)
     end # calc_contribution

data/lib/fselector/algo_discrete/Accuracy.rb

@@ -17,8 +17,12 @@ module FSelector
     def calc_contribution(f)
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
+        n = a+b+c+d
         
-        s = (a+d) / (a+b+c+d)
+        s = 0.0
+        if not n.zero?
+          s = (a+d) / n
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/AccuracyBalanced.rb

@@ -18,7 +18,10 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        s = (a/(a+c) - b/(b+d)).abs
+        s = 0.0        
+        if not (a+c).zero? and not (b+d).zero?
+          s = (a/(a+c) - b/(b+d)).abs
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/BiNormalSeparation.rb

@@ -7,8 +7,8 @@ module FSelector
 #
 #     BNS = |F'(tpr) - F'(fpr)|
 #      
-#     where F' is normal inverse cumulative distribution function
-#     R executable is required to calculate qnorm, i.e. F'(x)
+#     where F'(x) is normal inverse cumulative distribution function
+#     R equivalent: qnorm
 # 
 # ref: [An extensive empirical study of feature selection metrics for text classification](http://dl.acm.org/citation.cfm?id=944974) and [Rubystats](http://rubystats.rubyforge.org)
 #
@@ -25,8 +25,11 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        tpr, fpr = a/(a+c), b/(b+d)
-        s = (@nd.get_icdf(tpr) - @nd.get_icdf(fpr)).abs
+        s = 0.0
+        if not (a+c).zero? and not (b+d).zero?
+          tpr, fpr = a/(a+c), b/(b+d)
+          s = (@nd.get_icdf(tpr) - @nd.get_icdf(fpr)).abs
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/ChiSquaredTest.rb

@@ -22,12 +22,12 @@ module FSelector
     #
     # new()
     #
-    # @param [Boolean] correction Yates's continuity correction  
-    #   :yates, Yates's continuity correction
+    # @param [Boolean] correction Yates's continuity correction?  
+    #   no correction if nil, correction otherwise
     #
     def initialize(correction=nil, data=nil)
       super(data)
-      @correction = (correction==:yates) ? true : false
+      @correction = (correction || false)
     end
      
     
@@ -44,12 +44,16 @@ module FSelector
             "Chi-squared approximation may be incorrect"
         end
         
-        if not @correction
-          s = n * ((a*d-b*c)**2) /
-            (a+b) / (c+d) / (a+c) / (b+d)
-        else
-          s = n * (((a*d-b*c).abs - n/2))**2 /
-            (a+b) / (c+d) / (a+c) / (b+d)
+        s = 0.0
+        if not (a+b).zero? and not (c+d).zero? and 
+           not (a+c).zero? and not (b+d).zero?
+          if not @correction
+            s = n * ((a*d-b*c)**2) /
+               (a+b) / (c+d) / (a+c) / (b+d)
+          else
+            s = n * (((a*d-b*c).abs - n/2))**2 /
+               (a+b) / (c+d) / (a+c) / (b+d)
+          end
         end
         
         set_feature_score(f, k, s)

data/lib/fselector/algo_discrete/CorrelationCoefficient.rb

@@ -21,9 +21,12 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         n = a+b+c+d
-          
-        s = Math.sqrt(n) * (a*d-b*c) /
-          Math.sqrt( (a+b) * (c+d) * (a+c) * (b+d) )
+        
+        s = 0.0
+        if not ((a+b)*(c+d)*(a+c)*(b+d)).zero?  
+          s = Math.sqrt(n) * (a*d-b*c) /
+              Math.sqrt( (a+b) * (c+d) * (a+c) * (b+d) )
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/F1Measure.rb

@@ -24,7 +24,10 @@ module FSelector
       each_class do |k|
         a, b, c = get_A(f, k), get_B(f, k), get_C(f, k)
       
-        s = 2*a / (a+c+a+b)
+        s = 0.0
+        if not (a+c+a+b).zero?
+          s = 2*a / (a+c+a+b)
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/GMean.rb

@@ -20,7 +20,10 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        s = Math.sqrt( (a*d)/((a+c)*(b+d)) )
+        s = 0.0
+        if not ((a+c)*(b+d)).zero?
+          s = Math.sqrt( (a*d)/((a+c)*(b+d)) )
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/GSSCoefficient.rb

@@ -25,7 +25,10 @@ module FSelector
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         n = a+b+c+d
         
-        s = a/n * d/n - b/n * c/n
+        s = 0.0
+        if not n.zero?
+          s = a/n * d/n - b/n * c/n
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/GiniIndex.rb

@@ -22,11 +22,12 @@ module FSelector
       
       each_class do |k|
         a, b = get_A(f, k), get_B(f, k)
-        s += (a/(a+b))**2
+        s += (a/(a+b))**2 if not (a+b).zero?
       end
       
       # note: we've intentionally negated it
       s = s-1
+      s = -0.5 if s.zero? # Gini(max) = 0.5
       
       set_feature_score(f, :BEST, s)
     end # calc_contribution

data/lib/fselector/algo_discrete/MatthewsCorrelationCoefficient.rb

@@ -23,9 +23,11 @@ module FSelector
     def calc_contribution(f)
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
-        n = a+b+c+d
-      
-        s = (a*d-b*c) / Math.sqrt((a+b)*(a+c)*(b+d)*(c+d))
+        
+        s = 0.0
+        if not ((a+b)*(a+c)*(b+d)*(c+d)).zero?     
+          s = (a*d-b*c) / Math.sqrt((a+b)*(a+c)*(b+d)*(c+d))
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/McNemarsTest.rb

@@ -2,10 +2,10 @@
 # FSelector: a Ruby gem for feature selection and ranking
 #
 module FSelector
-# McNemar's test (MN), based on Chi-Squared test
+# McNemar's test (MNT), based on Chi-Squared test
 #
 #                 (B-C)^2
-#     MN(f, c) = ---------
+#     MNT(f, c) = ---------
 #                  B+C
 #
 # suitable for large samples and B+C >= 25
@@ -16,8 +16,8 @@ module FSelector
     #
     # new()
     #
-    # @param [Boolean] correction correction Yates's continuity correction  
-    #   :yates, Yates's continuity correction
+    # @param [Boolean] correction Yates's continuity correction?  
+    #   no correction if nil, correction otherwise
     #
     def initialize(correction=nil, data=nil)
       super(data)
@@ -36,10 +36,13 @@ module FSelector
             "Chi-squared approximation may be incorrect"
         end
         
-        if not @correction
-          s = (b-c)**2 / (b+c)
-        else
-          s = ((b-c).abs-0.5)**2 / (b+c)
+        s = 0.0
+        if not (b+c).zero?
+          if not @correction
+            s = (b-c)**2 / (b+c)
+          else
+            s = ((b-c).abs-0.5)**2 / (b+c)
+          end
         end
         
         set_feature_score(f, k, s)

data/lib/fselector/algo_discrete/MutualInformation.rb

@@ -25,7 +25,10 @@ module FSelector
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         n = a+b+c+d
         
-        s = Math.log2(a*n/(a+b)/(a+c))
+        s = 0.0
+        if not ((a+b)*(a+c)).zero?
+          s = Math.log2(a*n/(a+b)/(a+c))
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/OddsRatio.rb

@@ -24,8 +24,11 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        s = (a*d) / (b*c)
-        
+        s = 0.0
+        if not (b*c).zero?
+          s = (a*d) / (b*c)
+        end
+            
         set_feature_score(f, k, s)
       end
     end # calc_contribution

data/lib/fselector/algo_discrete/OddsRatioNumerator.rb

@@ -22,7 +22,10 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        s = a*d/(a+c)/(b+d)
+        s = 0.0
+        if not ((a+c)*(b+d)).zero?
+          s = a*d/(a+c)/(b+d)
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/Power.rb

@@ -31,7 +31,10 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        s = (d/(b+d))**(@k) - (c/(a+c))**(@k)
+        s = 0.0
+        if not (b+d).zero? and not (a+c).zero?
+          s = (d/(b+d))**(@k) - (c/(a+c))**(@k)
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/Precision.rb

@@ -18,7 +18,10 @@ module FSelector
       each_class do |k|
         a, b = get_A(f, k), get_B(f, k)
         
-        s = a/(a+b)
+        s = 0.0
+        if not (a+b).zero?
+          s = a/(a+b)
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/ProbabilityRatio.rb

@@ -22,7 +22,10 @@ module FSelector
       each_class do |k|
         a, b, c, d = get_A(f, k), get_B(f, k), get_C(f, k), get_D(f, k)
         
-        s = a * (b+d) / (a+c) / b
+        s = 0.0
+        if not (a+c).zero? and not b.zero?
+          s = a * (b+d) / (a+c) / b
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/Sensitivity.rb

@@ -18,7 +18,10 @@ module FSelector
       each_class do |k|
         a, c = get_A(f, k), get_C(f, k)
         
-        s = a/(a+c)
+        s =0.0
+        if not (a+c).zero?
+          s = a/(a+c)
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/algo_discrete/Specificity.rb

@@ -18,7 +18,10 @@ module FSelector
       each_class do |k|
         b, d = get_B(f, k), get_D(f, k)
         
-        s = d/(b+d)
+        s = 0.0
+        if not (b+d).zero?
+          s = d/(b+d)
+        end
         
         set_feature_score(f, k, s)
       end

data/lib/fselector/replace_missing_values.rb

@@ -11,7 +11,7 @@ module ReplaceMissingValues
     each_sample do |k, s|
       each_feature do |f|
         if not s.has_key? f
-          s[f] = my_value
+          s[f] = val
         end
       end
     end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: fselector
 version: !ruby/object:Gem::Version
-  version: 0.3.1
+  version: 0.4.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-04-03 00:00:00.000000000 Z
+date: 2012-04-04 00:00:00.000000000 Z
 dependencies: []
 description: FSelector is a Ruby gem that aims to integrate various feature selection/ranking
   algorithms and related functions into one single package. Welcome to contact me
@@ -38,8 +38,6 @@ files:
 - lib/fselector/algo_base/base_Relief.rb
 - lib/fselector/algo_base/base_ReliefF.rb
 - lib/fselector/algo_continuous/CFS_c.rb
-- lib/fselector/algo_continuous/discretizer.rb
-- lib/fselector/algo_continuous/normalizer.rb
 - lib/fselector/algo_continuous/PMetric.rb
 - lib/fselector/algo_continuous/ReliefF_c.rb
 - lib/fselector/algo_continuous/Relief_c.rb
@@ -72,9 +70,11 @@ files:
 - lib/fselector/algo_discrete/Sensitivity.rb
 - lib/fselector/algo_discrete/Specificity.rb
 - lib/fselector/algo_discrete/SymmetricalUncertainty.rb
+- lib/fselector/discretizer.rb
 - lib/fselector/ensemble.rb
 - lib/fselector/entropy.rb
 - lib/fselector/fileio.rb
+- lib/fselector/normalizer.rb
 - lib/fselector/replace_missing_values.rb
 - lib/fselector/util.rb
 - lib/fselector.rb