Theory of dynamic advising has been converted into universal tool based on dynamic normal mixtures used as environment model and on fully probabilistic design used as constructor of advising strategies. The resulting system has been applied in such diverse field as operating of rolling mills and supporting of medical doctors curing cancer of thyroid gland. Other applications are addressed, too.
The only way to solve the problem of bad traffic situation in cities is the application of signal traffic control. Daily traffic course and the requirements of public transport preferences strongly need the application of some intelligent traffic system. Such a system should provide some efficient traffic control using on-line adaptive methods.
Complex physical modeling, need for fast on-line computations on a large modeled time-spatial domain and sparse data are dominant characteristics of this important applied project.
These applications while being important on their own help us to deal with problems relying on prior information, see
lymphoscintigraphy, and careful modeling, see
nuclear. All of them including
internal test modeling capabilities of mixtures especially in cases when the number of learning data is small.
