The role of data mining in crime analysis is to detect all types of relevant structures in a dataset. Once filtered, those structures may point to previously unseen criminal activities. Data mining covers a wide range of methods and techniques whose potential strongly depends on the available dataset and the nature of the activity from which it is derived. Determining the most promising techniques to be applied in regards to different available forensic case databases is the aim of the research program.
A specific application to illicit drug profiling has been assessed. Analytical laboratory techniques are systematically applied to samples of cocaine seized by the police in order to extract their chemical profile. The data obtained is collated within a database. Cutting agents found are of particular interest, because they result from a treatment occurring toward the end of the distribution process. Their interpretation may then provide information on possible local illicit traffic networks. By focusing on the co-occurrence of set of cutting agents, relevant patterns are detected. This combinatorial approach using graph theory will be further tested on other crime data.