Integrate Citation detector with Metrics

app/models/metrics/algorithms.rb

@@ -48,8 +48,9 @@ def generate(month = nil)
                 else
                   count_matches(SearchEvent.includes(:term))
                 end
-      Metrics::Algorithms.create(month:, doi: matches[:doi], issn: matches[:issn], isbn: matches[:isbn],
-                                 lcsh: matches[:lcsh], pmid: matches[:pmid], journal_exact: matches[:journal_exact],
+      Metrics::Algorithms.create(month:, citation: matches[:citation], doi: matches[:doi], issn: matches[:issn],
+                                 isbn: matches[:isbn], lcsh: matches[:lcsh], pmid: matches[:pmid],
+                                 journal_exact: matches[:journal_exact],
                                  suggested_resource_exact: matches[:suggested_resource_exact],
                                  unmatched: matches[:unmatched])
     end
@@ -78,11 +79,25 @@ def count_matches(events)
     # @return does not return anything (the same matches Hash is passed in each loop but not explicitly sent back)
     def event_matches(event, matches)
       ids = match_standard_identifiers(event, matches)
+      citation = match_citations(event, matches)
       journal_exact = process_journals(event, matches)
       suggested_resource_exact = process_suggested_resources(event, matches)
       lcshs = match_lcsh(event, matches)
 
-      matches[:unmatched] += 1 if ids.detections.blank? && lcshs.detections.blank? && journal_exact.count.zero? && suggested_resource_exact.count.zero?
+      matches[:unmatched] += 1 if ids.detections.blank? && !citation.detection? && lcshs.detections.blank? && journal_exact.count.zero? && suggested_resource_exact.count.zero?
+    end
+
+    # Checks for Citation matches
+    #
+    # @param event [SearchEvent] an individual search event to check for matches
+    # @param matches [Hash] a Hash that keeps track of how many of each algorithm we match, to which this method will
+    #                       add.
+    # @return [Boolean] an indication of whether a citation was detector or not.
+    def match_citations(event, matches)
+      result = Detector::Citation.new(event.term.phrase)
+
+      matches[:citation] += 1 if result.detection?
+      result
     end
 
     # Checks for LCSH matches

db/migrate/20241011181823_add_citation_to_metrics_algorithms.rb

@@ -0,0 +1,5 @@
+class AddCitationToMetricsAlgorithms < ActiveRecord::Migration[7.1]
+  def change
+    add_column :metrics_algorithms, :citation, :integer
+  end
+end

test/fixtures/search_events.yml

@@ -22,6 +22,9 @@ old_month_pmid:
   term: pmid_38908367
   source: test
   created_at: <%= 1.year.ago %>
+current_month_citation:
+  term: citation
+  source: test
 current_month_issn:
   term: issn_1075_8623
   source: test

test/fixtures/terms.yml

@@ -27,7 +27,7 @@ doi:
   phrase: '10.1016/j.physio.2010.12.004'
 
 isbn_9781319145446:
-  phrase: 'Sadava, D. E., D. M. Hillis, et al. Life: The Science of Biology. 11th ed. W. H. Freeman, 2016. ISBN: 9781319145446'
+  phrase: 'Sadava, D. E., D. M. Hillis, et al. Life The Science of Biology. 11th ed. W. H. Freeman, 2016. ISBN: 9781319145446'
 
 journal_nature_medicine:
   phrase: 'nature medicine'

test/models/metrics/algorithms_test.rb

@@ -21,6 +21,12 @@
 
 class Algorithms < ActiveSupport::TestCase
   # Monthlies
+  test 'citation counts are included in monthly aggregation' do
+    aggregate = Metrics::Algorithms.new.generate(DateTime.now)
+
+    assert_equal 1, aggregate.citation
+  end
+
   test 'dois counts are included in monthly aggregation' do
     aggregate = Metrics::Algorithms.new.generate(DateTime.now)
 
@@ -85,6 +91,11 @@ class Algorithms < ActiveSupport::TestCase
     # drop all searchevents to make math easier and minimize fragility over time as more fixtures are created
     SearchEvent.delete_all
 
+    citation_expected_count = rand(1...100)
+    citation_expected_count.times do
+      SearchEvent.create(term: terms(:citation), source: 'test')
+    end
+
     doi_expected_count = rand(1...100)
     doi_expected_count.times do
       SearchEvent.create(term: terms(:doi), source: 'test')
@@ -117,6 +128,7 @@ class Algorithms < ActiveSupport::TestCase
 
     aggregate = Metrics::Algorithms.new.generate(DateTime.now)
 
+    assert_equal citation_expected_count, aggregate.citation
     assert_equal doi_expected_count, aggregate.doi
     assert_equal issn_expected_count, aggregate.issn
     assert_equal isbn_expected_count, aggregate.isbn
@@ -126,6 +138,12 @@ class Algorithms < ActiveSupport::TestCase
   end
 
   # Total
+  test 'citation counts are included in total aggregation' do
+    aggregate = Metrics::Algorithms.new.generate
+
+    assert_equal 1, aggregate.citation
+  end
+
   test 'dois counts are included in total aggregation' do
     aggregate = Metrics::Algorithms.new.generate
 
@@ -178,6 +196,11 @@ class Algorithms < ActiveSupport::TestCase
     # drop all searchevents to make math easier and minimize fragility over time as more fixtures are created
     SearchEvent.delete_all
 
+    citation_expected_count = rand(1...100)
+    citation_expected_count.times do
+      SearchEvent.create(term: terms(:citation), source: 'test')
+    end
+
     doi_expected_count = rand(1...100)
     doi_expected_count.times do
       SearchEvent.create(term: terms(:doi), source: 'test')
@@ -215,6 +238,7 @@ class Algorithms < ActiveSupport::TestCase
 
     aggregate = Metrics::Algorithms.new.generate
 
+    assert_equal citation_expected_count, aggregate.citation
     assert_equal doi_expected_count, aggregate.doi
     assert_equal issn_expected_count, aggregate.issn
     assert_equal isbn_expected_count, aggregate.isbn