I've used SPSS decision tree models to prioritise sewer cleansing to reduce rare but catastrophic blockages on a large sewer network. We're now collecting CCTV information on the high risk sewers and cleaning them. We have some poor and some good information from this work but only for the high risk sewers. We want to improve the risk scores as we go. This seems like an example of a Bayesian search and we could do some updating of the conditional probabilities. I also want to re-model using the new information and feed this back into the conditional probabilities.
Do you have any advice on the best way to structure this problem and choose a methodology?
Thanks in advance,
You may leave the list at any time by sending the command
to [log in to unmask], leaving the subject line blank.