Improve empty word handling

Change-Id: Ic202c5e7e0ef2682060aa77ac9821f87a960bdb0
diff --git a/systems/parse_spacy_pipe.py b/systems/parse_spacy_pipe.py
index f6dd7fa..0abfb33 100644
--- a/systems/parse_spacy_pipe.py
+++ b/systems/parse_spacy_pipe.py
@@ -128,9 +128,9 @@
 		words = text.split(' ')
 		# Filter out empty strings to avoid spaCy errors
 		words = [w for w in words if w]
-		# Handle edge case of empty input
+		# Handle edge case of empty input - use a placeholder token
 		if not words:
-			words = ['']
+			words = ['_EMPTY_']
 		# All tokens 'own' a subsequent space character in this tokenizer
 		spaces = [True] * len(words)
 		return Doc(self.vocab, words=words, spaces=spaces)