Advanced clustering methods for mining chemical databases in forensic science

摘要: Heroin and cocaine gas chromatography data are analyzed using several clustering techniques. A database with clusters confirmed by police investigation is used to assess the potential of analysis chemical signature these drugs in process. Results compared standard methods field drug profiling show that conventional approaches miss inherent structure data, which highlighted such as spectral its variants. Also, an approach based on genetic programming presented order tune affinity matrix algorithm. indicate all algorithms a quite different behavior two datasets, but both cases, exhibits level clustering, since there at least one type algorithm performs significantly better than chance. This confirms relevancy databases process understanding illicit market, information regarding trafficking networks can likely be extracted from composition drugs.