BayesFusion Support Forum

It is hard to say what exactly is happening in your environment but here is the address of the page with Microsoft error codes: https://msdn.microsoft.com/en-us/library/windows/desktop/ms681388(v=vs.85).aspx It seems that your error (500) is: ERROR_USER_PROFILE_LOAD 500 (0x1F4) User profile cannot b...

Hi Alessandro, Arc properties that have to do with their strengths are part of GeNIe and not SMILE, so you cannot find this information through SMILE. We may give users access to these properties through SMILE at some point, so please stay tuned. There is no difference in terms of SMILE classes and ...

You cannot place an equation in a CPT cell in GeNIe. However, you can freely copy and paste tables or their fragments between GeNIe and a program like Excel, so you can easily construct a CPT in Excel and then paste it into GeNIe. You can also use continuous nodes with equations that describe the in...

Dear Viliop, I'm not sure I understand your question, so let me try to answer what I do understand. Discrete variables are generally approximations of continuous variables. Whenever precision is not at stake or the variable can be naturally expressed in discrete terms, I would suggest discrete varia...

Virtual evidence implemented in GeNIe/SMILE is a standard calculation that you can read about in the literature. Here is one place where you can find information: https://www2.ee.washington.edu/techsite/papers/documents/UWEETR-2004-0016.pdf. Essentially, virtual evidence corresponds to a child node ...

Hi Viliop, I will be glad to help but please help me to understand your question. Are you referring to hybrid networks and a continuous node in such a network? What do you mean by a "final variable node"? A node with no successors? The number of samples can be set in Network Properties, Sampling tab...

Hi Laura, I'm afraid the only thing that you can copy and paste from the dialog are the tornado diagrams. Just right-click on the diagram and then select Copy. You can paste the copied diagrams in a program like Word. If you want access to the other numbers, like derivatives in the sensitivity calcu...

No, we don't have it yet but do stay tuned!

I'm afraid SMILE and GeNIe do not have general learning procedures for structural equation models. If you mean systems of linear equations with Gaussian noise, then GeNIe and SMILE do learn equations that you can interpret as structural equation models. I hope this helps.

Marek

Bayesian classifier and Bayesian network are two related concepts but they mean different things. Bayesian classifier is a model based on Bayesian methods aiming at classification. Bayesian networks are great classifiers. You have described a naive Bayes (class is the only parent of every feature no...

When answering your query, I would like to stress the idea that Bayesian networks are not about probabilities, they are about values. Any node in a Bayesian network is a variable, with well defined domain, whether it consists of a set of categories, discrete, or continuous values. Probabilities are ...

Meghan, Thanks for finding this bug -- we will fix it in the next release. I just wanted to comment here that having 365 states is not a usual modeling practice. I don't know your problem and can't advise without knowing it but I feel that you are asking for problems by having a node with so many st...

Hi Dana, SMILE API allows you to create and manage multiple networks at the same time as well as combine various sources of information in a single network, so I cannot see any problems with what you want to do. We do not support R directly as of yet but you can use SMILE in R through our Java wrapp...

GeNIe does not have any special facilities for GIS connectivity. However, GeNIe is just a simple GUI to SMILE, our inference engine, which is a general purpose library for decision-theoretic methods and can be linked to anything you want. SMILE API is accessible from C++, Java, .NET and Python. I kn...

There seems to be nothing wrong with your data, so I suspect that it is just confusion on your part. Discretization will obviously lead to loss of information and may decrease the quality of your model. Sometimes, however, you have no choice and need to discretize. I would definitely make then 2 for...