From 4ebd80fee52c04bd36781f846eae60019918345d Mon Sep 17 00:00:00 2001
From: Christoph Helma <helma@in-silico.ch>
Date: Thu, 21 Apr 2016 14:29:23 +0200
Subject: initial classification probabilities

---
 lib/classification.rb | 38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)

(limited to 'lib/classification.rb')

diff --git a/lib/classification.rb b/lib/classification.rb
index 0202940..b9b66f0 100644
--- a/lib/classification.rb
+++ b/lib/classification.rb
@@ -5,28 +5,28 @@ module OpenTox
 
       def self.weighted_majority_vote compound, params
         neighbors = params[:neighbors]
-        weighted_sum = {}
-        sim_sum = 0.0
-        confidence = 0.0
-        neighbors.each do |row|
-          sim = row["tanimoto"]
-          row["features"][params[:prediction_feature_id].to_s].each do |act|
-            weighted_sum[act] ||= 0
-            weighted_sum[act] += sim
+        feature_id = params[:prediction_feature_id].to_s
+        sims = {}
+        neighbors.each do |n|
+          sim = n["tanimoto"]
+          n["features"][feature_id].each do |act|
+            sims[act] ||= []
+            sims[act] << sim
+            #sims[act] << 0.5*sim+0.5 # scale to 1-0.5 
           end
         end
-        case weighted_sum.size
-        when 1
-          return {:value => weighted_sum.keys.first, :confidence => weighted_sum.values.first/neighbors.size.abs}
-        when 2
-          sim_sum = weighted_sum[weighted_sum.keys[0]]
-          sim_sum -= weighted_sum[weighted_sum.keys[1]]
-          sim_sum > 0 ? prediction = weighted_sum.keys[0] : prediction = weighted_sum.keys[1] 
-          confidence = (sim_sum/neighbors.size).abs 
-          return {:value => prediction,:confidence => confidence}
-        else
-          bad_request_error "Cannot predict more than 2 classes, multinomial classifications is not yet implemented. Received classes were: '#{weighted.sum.keys}'"
+        sim_all = sims.collect{|a,s| s}.flatten
+        sim_sum = sim_all.sum
+        sim_max = sim_all.max
+        probabilities = {}
+        sims.each do |a,s|
+          probabilities[a] = s.sum/sim_sum
         end
+        probabilities = probabilities.collect{|a,p| [a,sim_max*p]}.to_h
+        p_max = probabilities.collect{|a,p| p}.max
+        prediction = probabilities.key(p_max)
+        {:value => prediction,:probabilities => probabilities}
+
       end
     end
   end
-- 
cgit v1.2.3