Add Two-Parameter IRT Model and Corresponding Tests

- Implement Two-Parameter Model (2PL) for Item Response Theory
- Add tests for Two-Parameter Model
- Update main library file to include the new model
- Adjust README for the new model usage

README.md

@@ -1,7 +1,7 @@
 
 # IrtRuby
 
-IrtRuby is a Ruby gem that provides a simple implementation of the Rasch model for Item Response Theory (IRT). It allows you to estimate the abilities of individuals and the difficulties of items based on their responses to a set of items.
+IrtRuby is a Ruby gem that provides implementations of the Rasch model and the Two-Parameter model for Item Response Theory (IRT). It allows you to estimate the abilities of individuals and the difficulties of items based on their responses to a set of items.
 
 ## Installation
 

lib/irt_ruby.rb

@@ -2,6 +2,7 @@
 
 require "irt_ruby/version"
 require "irt_ruby/rasch_model"
+require "irt_ruby/two_parameter_model"
 
 module IrtRuby
   class Error < StandardError; end

lib/irt_ruby/two_parameter_model.rb

@@ -0,0 +1,60 @@
+# frozen_string_literal: true
+
+require "matrix"
+
+module IrtRuby
+  # A class representing the Two-Parameter model for Item Response Theory.
+  class TwoParameterModel
+    def initialize(data)
+      @data = data
+      @abilities = Array.new(data.row_count) { rand }
+      @difficulties = Array.new(data.column_count) { rand }
+      @discriminations = Array.new(data.column_count) { rand }
+      @max_iter = 1000
+      @tolerance = 1e-6
+    end
+
+    def sigmoid(x)
+      1.0 / (1.0 + Math.exp(-x))
+    end
+
+    def likelihood
+      likelihood = 0
+      @data.row_vectors.each_with_index do |row, i|
+        row.to_a.each_with_index do |response, j|
+          prob = sigmoid(@discriminations[j] * (@abilities[i] - @difficulties[j]))
+          if response == 1
+            likelihood += Math.log(prob)
+          elsif response.zero?
+            likelihood += Math.log(1 - prob)
+          end
+        end
+      end
+      likelihood
+    end
+
+    def update_parameters
+      last_likelihood = likelihood
+      @max_iter.times do |_iter|
+        @data.row_vectors.each_with_index do |row, i|
+          row.to_a.each_with_index do |response, j|
+            prob = sigmoid(@discriminations[j] * (@abilities[i] - @difficulties[j]))
+            error = response - prob
+            @abilities[i] += 0.01 * error * @discriminations[j]
+            @difficulties[j] -= 0.01 * error * @discriminations[j]
+            @discriminations[j] += 0.01 * error * (@abilities[i] - @difficulties[j])
+          end
+        end
+        current_likelihood = likelihood
+        break if (last_likelihood - current_likelihood).abs < @tolerance
+
+        last_likelihood = current_likelihood
+      end
+    end
+
+    def fit
+      update_parameters
+      { abilities: @abilities, difficulties: @difficulties, discriminations: @discriminations }
+    end
+  end
+end

spec/irt_ruby/rasch_model_spec.rb

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
-require "irt_ruby"
+require "spec_helper"
 
 RSpec.describe IrtRuby::RaschModel do
   let(:data) { Matrix[[1, 0, 1], [0, 1, 0], [1, 1, 1]] }

spec/irt_ruby/two_parameter_model_spec.rb

@@ -0,0 +1,33 @@
+require "spec_helper"
+
+RSpec.describe IrtRuby::TwoParameterModel do
+  let(:data) { Matrix[[1, 0, 1], [0, 1, 0], [1, 1, 1]] }
+  let(:model) { IrtRuby::TwoParameterModel.new(data) }
+
+  describe "#initialize" do
+    it "initializes with data" do
+      expect(model.instance_variable_get(:@data)).to eq(data)
+    end
+  end
+
+  describe "#sigmoid" do
+    it "calculates the sigmoid function" do
+      expect(model.sigmoid(0)).to eq(0.5)
+    end
+  end
+
+  describe "#likelihood" do
+    it "calculates the likelihood of the data" do
+      expect(model.likelihood).to be_a(Float)
+    end
+  end
+
+  describe "#fit" do
+    it "fits the model and returns abilities, difficulties, and discriminations" do
+      result = model.fit
+      expect(result[:abilities].size).to eq(data.row_count)
+      expect(result[:difficulties].size).to eq(data.column_count)
+      expect(result[:discriminations].size).to eq(data.column_count)
+    end
+  end
+end