I am building onea classifier where I have data offrom two known classclasses, but I want to capture any new or "unknown" classclasses as well in my prediction. However, in my case, I do not have any examples forof the unknown classclasses as the data is a multidimensional time-series coming out of sensors, and itsit's difficult to create such unknown data.
What is the best way to handle such cases. Thanks!?