saving the score in the db

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "python.linting.pylintEnabled": false
+}

app.py

@@ -61,15 +61,15 @@ def run():
 
 @app.route('/start')
 def start_pipeline():
-    dbManager = DBManager("ske_db")
+    dbManager = DBManager("ske_db",1)
 
-    pipeline = Pipeline(dbManager,"./knowledge_extractor/data/expert_types.csv")
-    scores  = pipeline.run()
-
+    #pipeline = Pipeline(dbManager)
+    #scores  = pipeline.run()
+    _thread.start_new_thread(Pipeline, (dbManager,1))
     #pprint.pprint(fv["seeds"].head())
     #fv["candidates"].to_csv("cand_fv.csv")
     #fv["seeds"].to_csv("seed_fv.csv")
-    return scores
+    return render_template('redirect.html',title='Completed Request')
 
 if __name__ == '__main__':
     app.run()

knowledge_extractor/data/expert_types.csv

@@ -0,0 +1,8 @@
+[
+"http://dbpedia.org/ontology/Broadcaster",
+"http://dbpedia.org/ontology/Artist",
+"http://dbpedia.org/ontology/Magazine",
+"http://dbpedia.org/ontology/Model",
+"http://dbpedia.org/ontology/Organisation",
+"http://dbpedia.org/ontology/TelevisionShow"
+]

knowledge_extractor/pipeline.py

@@ -48,8 +48,8 @@ def computeSeedVectors(self,seeds):
 
         for seed in seeds:
             #computer array of mentioned entity
-            mentions[seed["_id"]] = ehe.getEntities(seed)
-            ast_mentions[seed["_id"]] = ast.getEntities(seed)
+            mentions[seed["handle"]] = ehe.getEntities(seed)
+            ast_mentions[seed["handle"]] = ast.getEntities(seed)
 
         space_ehe = self.createSpace(mentions)
         space_ast = self.createSpace(ast_mentions)
@@ -79,8 +79,8 @@ def computeCandidatesVectors(self,cands,space_ast,space_ehe):
 
         for cand in cands:
             #computer array of mentioned entity
-            mentions[cand["_id"]] = ehe.getEntities(cand)
-            ast_mentions[cand["_id"]] = ast.getEntities(cand)
+            mentions[cand["handle"]] = ehe.getEntities(cand)
+            ast_mentions[cand["handle"]] = ast.getEntities(cand)
 
 
         print("Creating feature vector for the candidates")
@@ -109,11 +109,13 @@ def run(self):
         centroid = self.createCentroid(feature_vectors["seeds"])
         centroid = centroid.values
 
-        scores = feature_vectors["candidates"].apply(lambda row: cosine(row,centroid),axis=1)
-        pprint.pprint(scores)
+        scores = feature_vectors["candidates"].apply(lambda row: 1-cosine(row,centroid),axis=1)
+
+        self.db.saveScores(scores)
+
         return scores
 
-    def __init__(self,db,expertFile):
+    def __init__(self,db,experiment_id):
         self.alfa=0.7
         self.db=db
         self.expertFile = [
@@ -124,3 +126,4 @@ def __init__(self,db,expertFile):
                 "http://dbpedia.org/ontology/Organisation",
                 "http://dbpedia.org/ontology/TelevisionShow"
                 ]
+        self.run()

utils/DBManager.py

@@ -17,7 +17,20 @@ def getMentions(self,query):
     def getMentionType(self,query):
         collection = "entity"
         return self.db[collection].find(query,{"types":1})
+
+    def saveScores(self,scores):
+        collection = "rankings"
+
+        for k,v in scores.items():
+            score = {
+                "handle":k,
+                "score":v,
+                "experiment_id":self.experiment_id
+            }
+            self.db[collection].insert(score)
+
 
-    def __init__(self,dbmane):    
+    def __init__(self,dbmane,experiment_id):    
         self.client = MongoClient()
         self.db = self.client[dbmane]
+        self.experiment_id = experiment_id