diff --git a/bin/infermapping b/bin/infermapping
index d580858..188273a 100644
--- a/bin/infermapping
+++ b/bin/infermapping
@@ -3,11 +3,11 @@ from argparse import ArgumentParser
 import logging
 from typing import List, Type
 
-from pyvis.network import Network
 
 from semanticlabeling.labeledcolumn import YetUnknownTypeColumn
 from semanticlabeling.labelinferencer import SemanticLabelInferencer
 from util.file import InputFile
+from util import graphbuilder, graphvisualizer
 from util.knowledgesource import KnowledgeSource
 
 logging.basicConfig(level=logging.INFO)
@@ -17,13 +17,15 @@ logger = logging.getLogger(__name__)
 def main(
         input_file_path: str,
         input_file_cls: Type[InputFile],
-        target_ontology_paths: List[str]
+        target_ontology_paths: List[str],
+        visualize: bool
 ):
     logger.info(
         f'Semantic label inferencing called with input file {input_file_path} '
         f'and target ontologies {" ".join(target_ontology_paths)}')
 
     ontologies = list([KnowledgeSource(path) for path in target_ontology_paths])
+
     # input_file holds a list of labeled columns:
     # (Pdb) pp(input_file.__dict__)
     # {'columns': [<semanticlabeling.labeledcolumn.IDColumn object at 0x12c737080>,
@@ -72,44 +74,20 @@ def main(
     #  )
     # ]
     input_file = input_file_cls(input_file_path=input_file_path)
+
     label_inferencer = SemanticLabelInferencer(input_file, ontologies)
+    graph = graphbuilder.build_from_label_inferencer(label_inferencer)
+
+    if visualize:
+        graphvisualizer.visualize(graph)
 
-    network = Network()
-    for knowledge_source in ontologies:
-        for column_name, labeled_column in knowledge_source.columns.items():
-            if not isinstance(labeled_column, YetUnknownTypeColumn):
-                node_label = str(labeled_column)
-                network.add_node(column_name, label=node_label, color='green')
-
-        for column_name, labeled_column in knowledge_source.columns.items():
-            if not isinstance(labeled_column, YetUnknownTypeColumn):
-                source_node_id = column_name
-                for link_name, targets in labeled_column.links.items():
-                    for target in targets:
-                        target_node_id = target.column_name
-                        network.add_edge(source_node_id, target_node_id, title=link_name)
-
-    for labeled_column in label_inferencer.get_labeled_columns():
-        node_label = str(labeled_column)
-        network.add_node(labeled_column.column_name, label=node_label, color='red')
-
-    for labeled_column in label_inferencer.get_labeled_columns():
-        source_node_id = labeled_column.column_name
-        for link_name, targets in labeled_column.links.items():
-            for target in targets:
-                target_node_id = target.column_name
-                network.add_edge(source_node_id, target_node_id, title=link_name)
-    import pdb; pdb.set_trace()
-
-    network.show('network.html', notebook=False, )
-
-    pass
     raise NotImplementedError()
 
 
 if __name__ == '__main__':
     arg_parser = ArgumentParser()
     arg_parser.add_argument('input_file')
+
     arg_parser.add_argument(
         '--filetype',
         default='csv',
@@ -117,6 +95,7 @@ if __name__ == '__main__':
     )
 
     arg_parser.add_argument('target_ontologies', nargs='+')
+    arg_parser.add_argument('--visualize', action='store_true')
 
     args = arg_parser.parse_args()
 
@@ -128,5 +107,6 @@ if __name__ == '__main__':
     main(
         input_file_path=input_file_path,
         input_file_cls=input_file_cls,
-        target_ontology_paths=target_ontology_paths
+        target_ontology_paths=target_ontology_paths,
+        visualize=args.visualize
     )