You are here: Resources > FIDIS Deliverables > Forensic Implications > D6.7c: Forensic Profiling > 
Intelligence, risk analysis, detection and surveillance from a forensic profiling perspective  Title:
 Emerging Profiling Technologies


Perspectives: virtual persons and forensic profiling

Virtual persons is a model defined in FIDIS which has already been presented  and is the subject of its own deliverable . It consists of abstract representations that may describe physical persons or groups of persons, computer software, as well as other virtual entities relevant in modern information societies. Associate to virtual persons are their profiles with their attributes, roles, abilities, acquisitions, habits, behaviours, etc.

This concept may be useful to connect crime/criminal intelligence analysis. Unknown offenders may be described as virtual persons that do not necessitate designating a particular physical person to be worked on, or may even point to several different physical individuals changing in the course of the investigation. For instance, virtual persons may be useful when considering serial crimes and the virtual offender(s) imagined through individual profiling (figure 3). Those structures may be scrutinised without necessitating permanent connection with physical persons; this connection may also change in the course of the investigation, according to a hypothetical-deductive reasoning process. From the profile, general knowledge about types of offenders may be used to refine the description of the offender. For instance, from the description of the person and his behaviour and general knowledge about typical patterns of mobility, some useful indications for localisation are inferred. Moreover, this profile may also apply to a particular data subject. For instance, if it is assumed that the individual is living in a particular region, the subject’s data concerning habitants are immediately linked.  

Figure : Description of a series of crimes using the concept of virtual persons. The physical and virtual worlds are separated. In the virtual world, perpetrators are not known, but their profiles can be worked on in order to point to a restricted set of potential candidates. In this example, the geographical distribution of the incidents may show a pattern typical to the behaviour of specific types of offenders.


This formalism may be even more suitable for modelling some confidence games that occur on the internet. Advance fee fraud scenario used by scammers to deceive gullible victims (see above) may be usefully described with virtual persons: persons appearing in the scenario and using different identities to deceive victims, as well as representations of the scammers themselves with their different roles in the fraud. This exercise should be the object of future use cases studies.  

However, it must be emphasised that the final objective of the investigation remains proving the existence of a crime, linking the activities to human beings, as well as collecting, preserving, interpreting and presenting evidence. Thus, the connection between virtual and physical worlds concentrates the attention. Using virtual persons within an investigative framework necessitates further research. It is not claimed here to have made a definitive account and even appropriate use of virtual persons, but only provide insight of how this notion may be linked to some aspects of forensic profiling and help provide the material that will help test the adequacy of the model in a forensic context through the use of cases studies. 

Forensic profiling follows various objectives related to the interpretation process, investigation or intelligence. Beyond difficulties and limits described above, the number of applications and ideas for potential developments increases rapidly. 

The following examples illustrate this variety by some emerging uses of forensic profiling. Some of these approaches also include the search for the application of suitable data mining technologies.  

The first example discuss the kind of information digital image may reveal about the used image acquisition device, as well as possible characteristics for interpreting the link between the trace and its putative source. The detection of images alterations belongs also to the whole forensic process presented.  

Another original illustration pertains to 3D-modelling of CCTV footage. Through the use of this technique, several parameters useful to reconstruct the activity recorded can be inferred, such as distances, locations of objects and persons, or their speed. In particular, this approach helps to test alternative hypothesis about scenarios. 

Emerging profiling and data mining technologies have still to be tested in order to situate better their potential in the analysis of large quantities of crime data, and to help provide timely and accurate intelligence. A centre of expertise on intelligent data analysis set-up in Holland for this purpose is presented.  

Finally, the use of forensic profiling for intelligence purposes is illustrated through an actual illicit-drug profiling process. Data mining technologies for detecting patterns in the chemical profile of the illicit substances seized have been tested. They help separate and detect changes in the organisation of the market. The harmonisation of methods used by several forensic laboratories is a condition to extend the use of the approach to an international level.  


Intelligence, risk analysis, detection and surveillance from a forensic profiling perspective  fidis-wp6-del6.7c.Forensic_Profiling.sxw  Emerging Profiling Technologies
11 / 27