Hi sec_ware,



I am quite grateful to you that you took this seriously.

But, I am still only a high school student and have never ever heard of Ford Fulkerson method

I know that you have quite a strong background in theoritical maths. But for me, no such thing at present.

1. Yes, "Bayesian Networks" are used to analyse such probabilities and they are quite succesful. You can feed any amount of causes and some predefined effects in a Bayesian Network and as a prequisite, define a few possible outcomes. After that as you use the system, it learns and provides accurate information and after a certain amount of usage it becomes impeccable. Bayesian Networks are not only used in connection to Bayes Theorem but also with other models like Hidden Markov Models.

2. My goal is to find an optimal path. Given in a LAN, the speed of transmission is quite high. So I don't think that figuring out a priori probabilities would not be such a mean task. I have heard of data transmission techniques called CSMA/CD. I have no experience of them and I was thinking if this type of algorithm can be used in such schemes. In this context, I would like to request the members to provide me some reading stuff and links.

3. I have used Bayes Theorem as I said Bayesian Networks learn by itself and if needed can also start from scratch. I think this is better than the multi-layered neural networks.

I would try to learn about the Ford method you have mentioned and then only can I map the algorithm to it.

4. Please excuse me for now. I have to go offline. Will be back within 24 hours........

And why would I take offence when you are posting