1
$\begingroup$

I want to ask if we have two graph one of them Bayesian Network and the other one just regular graph, how we can distinction between them.

$\endgroup$

closed as unclear what you're asking by Xi'an, John, COOLSerdash, whuber May 12 '15 at 20:28

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 1
    $\begingroup$ A Bayesian Network can be represented as a graph. $\endgroup$ – spdrnl May 12 '15 at 19:00
  • $\begingroup$ Could you be more specific about what a "regular graph" might be and how it would be presented or described to you? $\endgroup$ – whuber May 12 '15 at 20:28
  • 1
    $\begingroup$ @whuber I imagine "regular graph" in this case means "any other graph" as opposed to some technical term $\endgroup$ – shadowtalker May 12 '15 at 20:29
  • $\begingroup$ @ssde Thank you, but I would like the OP to explain what they mean rather than hearing your guess about what they might mean. Even "any other graph" is vague: are we talking about directed or undirected graphs? With or without labeled edges? Etc., etc. $\endgroup$ – whuber May 12 '15 at 20:32
0
$\begingroup$

Isn't Bayesian networks directed, with information flowing in only on direction and asymmetrical? Thus:

  1. if node i have a directed link to j, and j have a directed link to i then it ain't a Bayesian network,
  2. If the network ain't directed then it ain't a Bayesian network
  3. If you can start at a node and follow a path that loops you back to the node you started on then it ain't a Bayesian network.

I can only see test to falsify whether a graph qualifies as a Bayesian network or not. I don't think it I s possible to say for sure since any graph can have the same structure as a Bayesian network without being a Bayesian network. Whether a graph actually is a Bayesian network depends on the interpretation and the context (I.e. The code the network is processed by).

But if you have graph A and B. You know one of them is a Bayesian graph and the other not, then you can run the three tests above. If A fails, then B is the one and vice versa. If both A and B passes the test, then you have a problem.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.