# Bayesian Network and Samiam: Predicting Elections

#### zulfi100

Joined Jun 7, 2012
656
Hi,

In the diagram, I have shown 3 major issues but let’s consider that there are 4 issues and I have got following data related to Texas state:

C19=COVID19 = 38%=0.38

EJ= Economy and Job (28%)=0.28

R=Racism(9%)=0.09

HC = Health care = (9%)=0.09

Party Affiliations:

D= Democrat = 39%=0.39

Re =republican= 40%=0.04

N = neutral

There are total 17 million voters

Somebody please guide me how can I use the above data to predict the inclination of voters to one of the two candidates in a two party System where one of the candidates is the incumbent.Please feel free to suggest the modifications.
samiam

Zulfi.

Last edited:

#### nsaspook

Joined Aug 27, 2009
13,534
Is this a joke?

#### zulfi100

Joined Jun 7, 2012
656
Hi,
First try would look a joke, please guide me how to improve. I am trying to understand the bayesian network.

Zulfi.

#### nsaspook

Joined Aug 27, 2009
13,534

#### zulfi100

Joined Jun 7, 2012
656
Hi,

Zulfi.

#### nsaspook

Joined Aug 27, 2009
13,534

#### zulfi100

Joined Jun 7, 2012
656
Hi,
Thanks for other link. It provides some formulae. Can I use my network to calculate number of voters for a particular candidate using conditional probabilities?

Zulfi.

#### WBahn

Joined Mar 31, 2012
30,286
Hi,
Thanks for other link. It provides some formulae. Can I use my network to calculate number of voters for a particular candidate using conditional probabilities?

Zulfi.
No, because probabilities are just that -- there's a reason they are not called certainties.

#### zulfi100

Joined Jun 7, 2012
656
Hi,
Do you know how to use samiam tool?
Zulfi.

#### zulfi100

Joined Jun 7, 2012
656
Hi,
Can I use the probabilities to make a prediction that the candidate has got above 75% votes?

Zulfi.

#### WBahn

Joined Mar 31, 2012
30,286
Hi,
Can I use the probabilities to make a prediction that the candidate has got above 75% votes?

Zulfi.
I don't see how you could use the information in your first post to do so. You haven't given anything remotely like the conditional probabilities you would need to have.

#### zulfi100

Joined Jun 7, 2012
656
Hi,
I have normalized the percentages to values between 0 and 1, consider them as conditional probilities. But do you know how to use the tool samiam?

Zulfi.

#### WBahn

Joined Mar 31, 2012
30,286
Hi,
I have normalized the percentages to values between 0 and 1, consider them as conditional probilities. But do you know how to use the tool samiam?

Zulfi.
Conditional probabilities of what?

Do you even understand what a conditional probability is?

And, no, I do not how to use this particular tool for visualizing Bayesian networks, but there appear to be no shortage of tutorials available.

#### zulfi100

Joined Jun 7, 2012
656
Hi,
I am actually looking for somebody who knows the tool.
I have gone through those tutorials but I can't create any network, that's why I have posted here because its a ML forum. If you can guide me after reading the tutorial, I would check again in 10-15 minutes.

Zulfi.

#### WBahn

Joined Mar 31, 2012
30,286
I'm not going to read the tutorial to learn to use a tool that I have no desire to learn, let along install on my machine. You need to learn how to teach yourself instead of expecting to be spoonfed everything.

#### zulfi100

Joined Jun 7, 2012
656
I'm not going to read the tutorial to learn to use a tool that I have no desire to learn, let along install on my machine. You need to learn how to teach yourself instead of expecting to be spoonfed everything.
Okay then don't tell me about the tutorial. I have read it and I know that it does not solve the probem. You were telling me that, that's why I asked you to read it so that you know how much these tutorials are useful.

#### nsaspook

Joined Aug 27, 2009
13,534
Okay then don't tell me about the tutorial. I have read it and I know that it does not solve the probem. You were telling me that, that's why I asked you to read it so that you know how much these tutorials are useful.
Did you make any progress in predicting the past?