NCJ Number
141733
Journal
Journal of Forensic Sciences Volume: 38 Issue: 2 Dated: (March 1993) Pages: 292-301
Date Published
1993
Length
10 pages
Annotation
A personal computer was used to search a large database of cocaine signatures rapidly to identify database entries that closely resembled a given reference cocaine exhibit.
Abstract
Computer software developed jointly between the North Carolina State Bureau of Investigation (NCSBI) and the Center for Systems and Engineering of the Research Triangle Institute is used daily to aid forensic experts in identifying signatures that originate from the same batch. The program simulates the pattern recognition capability of the multilayer perceptron neural network to identify similar cocaine signatures. An ideal cocaine signature matching technique must maximize the probability of identifying signatures from the same batch while minimizing the probability of false positives. Using the neural network software, researchers conducted several experiments on cocaine signatures; all chromatographic signature patterns were correctly recognized and matched with no false positives or false negatives. Intelligence reports derived from these database searches have been used by undercover agents in the field who are working on investigations of drug-related conspiracies. 2 tables, 4 figures, and 21 references