Extract more contextual info from the words.

A property dictionary is now only to be considered equal to a word
when it shares at least one group, or neither has groups.

naive-nlu/tree_nlu/knowledge_base.py

@@ -22,7 +22,6 @@ class KnowledgeBase(object):
         knowledge_before = copy.deepcopy(self.knowledge)
 
         # Parse everything
-        parsed_examples = []
         for example in examples:
             # If there's parsed data, leverage it ASAP
             if 'parsed' in example:

naive-nlu/tree_nlu/knowledge_evaluation.py

@@ -273,3 +273,21 @@ def integrate_information(knowledge_base, example):
     return tagged_with_ast(
         ast, elements,
         knowledge_ingestion[method](knowledge_base, elements, *args))
+
+def can_be_used_in_place(knowledge, token, minisegment):
+    if token not in knowledge.knowledge:
+        return False
+
+    info = knowledge.knowledge[token]
+    info_groups = info.get('groups', set())
+    minisegment_groups = minisegment.get('groups', set())
+
+    # Common group
+    if len(info_groups & minisegment_groups) > 0:
+        return True
+
+    # Neither has a group
+    elif len(info_groups) == 0 == len(minisegment_groups):
+        return True
+
+    return False

naive-nlu/tree_nlu/parsing.py

@@ -395,8 +395,7 @@ def is_definite_minisegment(minisegment):
 
 def match_token(knowledge, next_token, minisegment):
     if isinstance(minisegment, dict):
-        # TODO: check if the dictionary matches the values
+        return knowledge_evaluation.can_be_used_in_place(knowledge, next_token, minisegment)
-        return True
     elif isinstance(minisegment, str):
         # TODO: check if the two elements can be used in each other place
         return next_token == minisegment

naive-nlu/tree_nlu/tests/basic.py

@@ -98,6 +98,12 @@ base_knowledge = {
     'fly': {
         "groups": {'verb'},
     },
+    'bus': {
+        "groups": {'noun'},
+    },
+    'run': {
+        "groups": {'verb'},
+    },
     'swim': {
         "groups": {'verb'},
     },

naive-nlu/tree_nlu/tests/gac_100.py

@@ -1,3 +1,4 @@
+import logging
 from ..knowledge_base import KnowledgeBase
 from ..utils.visuals import show_progbar
 
@@ -109,7 +110,7 @@ examples = [
     # },
     ('text_example',
      {
-         "question": "Is milk white?",
+         "question": "is milk white?",
          "affirmation": "milk is white",
          "answer": True,
      }),
@@ -662,9 +663,15 @@ base_knowledge = {
     'planet': {
         "groups": {'noun', 'group'},
     },
+    'white': {
+        "groups": {'noun', 'color', 'concept', 'property'},
+    },
     'green': {
         "groups": {'noun', 'color', 'concept'},
     },
+    'milk': {
+        "groups": {'noun'},
+    },
     'fly': {
         "groups": {'verb'},
     },
@@ -713,11 +720,12 @@ def main():
         elif example_type == 'text_example':
             show_progbar(i, total, data['affirmation'])
             affirmation = data['affirmation']
+            logging.debug("Processing affirmation: {}".format(affirmation))
+            _, _, _ = knowledge.process(affirmation)
 
             show_progbar(i, total, data['question'])
             question = data['question']
-
+            logging.debug("Processing question : {}".format(question))
-            _, _, _ = knowledge.process(affirmation)
             result, _, _ = knowledge.process(question)
 
             if result != data['answer']: