Kommentare hinzugefügt

scripts/preprocessing/senseval/preprocess_senseval_method1.py

@@ -146,7 +146,9 @@ def get_lem_pos(word, d_no, sent_no, tb_docs):
 
 def preprocess(xml_file, tb_files):
 	"""
-
+	this function opens a senseval file 
+	returns a list containing a list for each document, each containing a list of sentences
+	containing a dictionary for every word to be disambiguated
 	"""
 	documents = read_xml(xml_file)
 	answer_ids = documents[1]
@@ -174,7 +176,7 @@ def preprocess(xml_file, tb_files):
 					lem_pos = get_lem_pos(word,i,j,tb_docs)
 					info['lemma'] = lem_pos[0]
 					info['pos'] = lem_pos[1]
-
+					# connects compound words per underscore
 					if ' sats=' in word:
 						for l, word2 in enumerate(sent):
 							id_m = id_p.search(word2).group()