Specimen identification rates in % (correctly identified/misidentified/not identified) for Dalbergia spp. using six different classification methods, for each of the seven barcodes.

TaxonDNA: Best close match results. Not identified rates are summed over the “Ambiguous” and “No match” categories, see S4 Table for details. BLOG: percentage correct classification for test file, using 90% slicing at species level. Naïve Bayes = Bayesian, SMO = Support vector machine, J48 = decision tree, Jrip = rulebased, all four classification methods in WEKA, tested with 10-fold crossvalidation. See materials and methods for details on the analyses.Specimen identification rates in % (correctly identified/misidentified/not identified) for Dalbergia spp. using six different classification methods, for each of the seven barcodes.