Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm trying to understand bayesan networks also I created a simple bayesan network according to same sample date.

This is the network (created with Hugin Lite)

enter image description here

There is one class (Failure) and two attributes (light and temperature).

Failures = kinds of failure (three possible values: Mechanic , Electric , None) Light = indicates if a light is on or off when there is a failure (which can be Mechanic OR Electric) Temperature = indicates the temperature (three possible values: low,normal,high)

Light and temperature are independent variables

In the beginning there is NO EVIDENCE about variables. These are the probability tables of "A PRIORI probability":

enter image description here

Now an EVIDENCE is provided. So we'll have a "A POSTERIORI Probability"

We know that:

  • Light is ON (so in the first table there will be just one row: Light = ON)
  • Temperature is LOW (so in the second tablet there will be just one row: Temperature = LOW)

The column of both table remain the same (Mechanic,Electric,None).

My question is: how should I update conditional table?

I'm sure there is some relationship with "Bayes formula" but I'm a bit confused.

This was an homework but I'll have to a similar (not the same) homework next month for an university project. My aim is to understand how to calculate the new probability.

Thank you in advance for any hint.

share|cite|improve this question

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.