Research objectives and hypothesis/research questions:
The assumption is that the solution in question, using the universal HMM method, allows its application in IO, in view of the need to select from the multitude of characterized data – metadata – patterns of conduct defined as a „negative impulse” (directional anomaly).

Research methods:
Reviewing the existing research on HMM issues and the possibilities of implementing this method for the purposes of combating crime, including money laundering and terrorism.

Main results:
Due to the lack of obtaining direct data on terrorist activities, their incompleteness and the need to predict the possibility of events that threaten public or financial security – there is a need to build knowledge about terrorists based on data not directly related to this type of crime (as indirect observational results). This also translates into the creation of training sets for mathematical models of counteracting. As a consequence, conducting research based on a small amount of data and available in larger time intervals may give inadequate results. In addition, the use of a probabilistic approach creates opportunities to design counteracting typified negative behaviors.

Implications for theory and practice:
The possibility of using mathematical models of counteraction in the scope of conducting reconnaissance analysis by obligated institutions and for the purposes of conducting analytical activities as part of the application of criminal analysis in the police and special services.