I try to explain my question with an example. Assume we have a network of 3 computers. We can measure different entity in this system such as CPU load, disk load, Log messages and etc. (all the entities are numeric). Let's the vector x holds all these entities(features). By running different scenario in the network, we are able to gather a number of data points which contain a feature vector x and a label y showing whether this scenario is a bad behaviour in the network or not. So, up to now we have a network and observe it for a while and all this information summarized in couple of data points.Let's call this network reference model.
Now, assume you are given a new network and you are able to run some experiments on that and gather data points like as before. Now the task is to evaluate how much this new model is similar to the reference model with respect to your measurements. Is it aligned with the reference model in terms of behaviour (Verify the model).
So, what are the available methods to approach above goal???