Filter out stop words

Remove stop words from automatically generated evidences.

Author: Marcin Kardas

sota_extractor2/models/linking/context_search.py

@@ -31,7 +31,7 @@ def __init__(self, taxonomy):
     @staticmethod
     def evidences_from_name(key):
         x = normalize_dataset_ws(key)
-        y = x.split()
+        y = [w for w in x.split() if w not in manual_dicts.stop_words]
         return [x] + y if len(y) > 1 else [x]
 
     @staticmethod

sota_extractor2/models/linking/manual_dicts.py

@@ -172,3 +172,11 @@
     'Top-1 Accuracy': 'Top-1 Error Rate',
     'Top-5 Accuracy': 'Top-5 Error',
 }
+
+stop_words = {
+    "a", "an", "and", "are", "as", "at", "be", "but", "by",
+    "for", "if", "in", "into", "is", "it",
+    "no", "not", "of", "on", "or", "such",
+    "that", "the", "their", "then", "there", "these",
+    "they", "this", "to", "was", "will", "with"
+}