initial parsing with nltk

tomviner · Jul 7, 2011 · f45693e · f45693e
1 parent 61507af
commit f45693e
Showing 1 changed file with 15 additions and 0 deletions.
diff --git a/parse.py b/parse.py
@@ -0,0 +1,15 @@
+import nltk
+
+def get_parts(text):
+    text = nltk.word_tokenize(text)
+    parts = nltk.pos_tag(text)
+    dic = {}
+    for work, part in parts:
+        if part in dic:
+            dic[part].append(work)
+        else:
+            dic[part] = [work]
+    return dic
+
+get_parts("hi how are you")
+