Pk. Hopke et Dl. Massart, REFERENCE DATA SETS FOR CHEMOMETRICAL METHODS TESTING, Chemometrics and intelligent laboratory systems, 19(1), 1993, pp. 35-41
There is a need for the chemometrics community to have sets of referen
ce data that have known or well characterized properties. These data s
ets can then be used to test the utility and efficiency of new methods
such that they can be compared with previous results on the same data
. A depository for such data sets has been established by the Editors
of Chemometrics and Intelligent Laboratory Systems with several data s
ets being the initial entries. These data include several sets for pat
tern recognition studies (Italian olive oils and human thyroid functio
n test results) and the simulated data sets created for testing models
that apportion airborne particulate matter mass to particle emission
sources (receptor models). The criteria and procedures for the additio
n of new data sets to the collection are presented along with the mech
anism for obtaining such data and restrictions on their use.