Discrimination of Marihuana Using Cluster Analysis

Syuji OKUYAMA and Toshiyuki MITSUI

Criminal Science Laboratory, Aichi Police H.Q., 2-1-1, Sannomaru, Naka-ku, Nagoya-shi 460, Japan

(Recieved : February 13, 1995; Accepted for publication : September 7, 1995)

Marihuana was discriminated using cluster analysis programing in Basic. The peak areas of three compounds in marihuana separated using gas chromatography (GC) and the areas of eight mass fragment ions obtained using gas chromatograph-mass spectrometry (GC-MS) were used for the discrimination. These areas were corrected using an internal standard and normalized according to a definite rule. This normalization method gave smaller experimental errors than those found by using only an internal standard. The normalization values were used for cluster analysis. As a result, the discrimination of marihuana was performed using the minimum Euclidean distance and the result was shown with the dendrogram obtained from the cluster analysis.

Keywords: Cluster analysis, Marihuana, Gas chromatography, Gas chromatograph mass-spectrometry, Discrimination


Abstract in Japanese

Text in English


Author Index
Keyword Index
Return