Merge pull request #37 from Knox-AAU/concept-linking

Concept linking

concept_linking/data/files/EvaluationData/Results/ML_DK.json → concept_linking/data/files/EvaluationData/Results/ML_DK_PC.json

@@ -36,7 +36,7 @@
             "knox-kb01.srv.aau.dk/Aalborgs",
             "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
             "http://dbpedia.org/ontology/Place",
-            0
+            1.0
         ]
     },
     {
@@ -140,15 +140,15 @@
             "knox-kb01.srv.aau.dk/Jonas_Wind",
             "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
             "http://dbpedia.org/ontology/Organisation",
-            1.0
+            0
         ]
     },
     {
         "FC København sikrede sig sejren mod Brøndby IF, da målscorerne Jonas Wind og Kasper Kusk viste imponerende præstationer.": [
             "knox-kb01.srv.aau.dk/Kasper_Kusk",
             "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
             "http://dbpedia.org/ontology/Organisation",
-            1.0
+            0
         ]
     },
     {

concept_linking/data/files/EvaluationData/Results/Prompt_EN.json → concept_linking/data/files/EvaluationData/Results/Prompt_EN_PC.json

@@ -4,8 +4,7 @@
             "knox-kb01.srv.aau.dk/Martin_Kjær",
             "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
             "http://dbpedia.org/ontology/Agent",
-            1,
-            "We have defined Agent as an entity that has agency. PersonFunction would also be correct here"
+            1
         ]
     },
     {

concept_linking/data/files/EvaluationData/Results/Untrained_EN.json → concept_linking/data/files/EvaluationData/Results/Untrained_EN_OC.json

@@ -84,7 +84,7 @@
             "knox-kb01.srv.aau.dk/Martin_Kjær",
             "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
             "https://dbpedia.org/ontology/Person",
-            1,0
+            1.0
         ]
     },
     {
@@ -332,7 +332,7 @@
             "knox-kb01.srv.aau.dk/Apple_Inc",
             "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
             "https://dbpedia.org/ontology/Organisation",
-            0
+            0.8
         ]
     },
     {

concept_linking/data/files/EvaluationData/evaluationSet_EN_small.json

@@ -19,17 +19,17 @@
                 ]
             },
             {
-                "sentence": "Michelle Obama is an person.",
+                "sentence": "Michelle Obama is a person.",
                 "sentenceStartIndex": 0,
-                "sentenceEndIndex": 24,
+                "sentenceEndIndex": 26,
                 "entityMentions": [
                     {
-                        "name": "Barack Obama",
+                        "name": "Michelle Obama",
                         "type": "Entity",
                         "label": "PERSON",
                         "startIndex": 0,
-                        "endIndex": 11,
-                        "iri": "knox-kb01.srv.aau.dk/Barack_Obama"
+                        "endIndex": 13,
+                        "iri": "knox-kb01.srv.aau.dk/Michelle_Obama"
                     }
                 ]
             }

concept_linking/main.py

@@ -14,10 +14,10 @@ def entity_type_classification(input_data):
     #stringComparisonSolution(input_data)
 
     # Untrained Spacy
-    #untrainedSpacySolution(input_data)
+    untrainedSpacySolution(input_data)
 
     # PromptEngineering
-    perform_entity_type_classification(input_data)
+    #perform_entity_type_classification(input_data)
 
     # Machine Learning
     #predict(input_data)

concept_linking/solutions/PromptEngineering/main.py

@@ -14,6 +14,8 @@
 # Local API url docker
 # api_url = "http://llama-cpu-server:5000/llama"
 
+
+
 # Remote API url
 # api_url = "http://knox-proxy01.srv.aau.dk/llama-api/llama"
 

concept_linking/solutions/StringComparison/utils.py

@@ -14,8 +14,6 @@ def translateWordToEn(word, language):
 
 
 from difflib import SequenceMatcher
-
-
 # returnerer en konstant mellem 0 og 1. 0 er et elendigt match, 1 er et eksakt match.
 def similar(a, b):
     return SequenceMatcher(None, a.lower(), b.lower()).ratio()

relation_extraction/relation_extractor.py

@@ -1,8 +1,9 @@
 # from relation_extraction.NaiveMVP.main import handle_relation_post_request
+from relation_extraction.NaiveMVP.main import handle_relation_post_request
 from relation_extraction.multilingual.main import begin_relation_extraction
 
 class RelationExtractor():
     @classmethod
     def begin_extraction(self, data):
-        # handle_relation_post_request(data) # Naive solution
+        handle_relation_post_request(data) # Naive solution
         begin_relation_extraction(data) # Multilingual solution

test/test_server/test_server.py

@@ -10,8 +10,8 @@ def setUp(self):
     # Valid: authorized and correct format
     @patch('os.getenv', return_value="some_api_secret")
     # @patch('concept_linking.main.stringComparisonSolution', return_value=Mock())
-    # @patch('concept_linking.main.untrainedSpacySolution', return_value=Mock())
-    @patch('concept_linking.main.perform_entity_type_classification', return_value=Mock())
+    @patch('concept_linking.main.untrainedSpacySolution', return_value=Mock())
+    # @patch('concept_linking.main.perform_entity_type_classification', return_value=Mock())
     # @patch('concept_linking.main.predict', return_value=Mock())
     @patch('relation_extraction.relation_extractor.RelationExtractor.begin_extraction', return_value=Mock())