Does anyone her use bayesian statistics or bayesian networks?

[adam_d]

Over the years I have been looking at predictive models for sports, mainly soccer/football and I am currently interested in bayesian networks. Unfortunately I only have a basic understanding so far and the literature available seems to be a bit advanced for me. I am currently looking into the methods for learning the structure and parameters of the network. My main interest in the bayesian networks is because I believe that combining subjective and objective data could be a substantial benefit when it comes to forecasting.... what do you guys think? any experiences with bayesian methods?

[knembo]

Re: Does anyone her use bayesian statistics or bayesian networks? Hi I have tried some machine learning alghoritm and bayesian network too using an open source software named WEKA.This software is simple to use. Basically you need to divide your input dataset in a so called training set containing all data you think could be good predictor (points,made and take goal etc.). You need to set the target of your prediction for example the results as home win draw or away win. The training data must contain the target of course. Then you have to apply your alghoritm, for example naive bayes on training data. In Weka there are lots of alghoritm available. In the end you have to apply your trained alghorithm on the matches you want to forecast. This is the very basic of classification problem. I am a newbie in this field and so far no profitable strategies found. You don't need to be very skilled in math and statistic to use the software. Of course it could be used as a black box but if you need fine tuning of the alghoritm you need to spends some time in learning. Regards