Additional file 1 of Machine learning approach to support taxonomic species discrimination based on helminth collections data

Additional file 1: Table S1. Morphometry: measurements of length, width, plug width, plug thickness and shell thickness; morphology: shell ornamentations, 1, smooth; 2, punctuated; 3, reticulated type I; 4, reticulated type II; host: fish, amphibian, reptile, avian, mammals; geographical location: South America, Central America, North America, Europe, Africa, Asia, Oceania. Response variable were 1 = yes or presence, 0 = no or absence, and -1 = no information. Table S2. Results of statistical analysis using AUC values performed on RStudio version 3.5.1. P-value of the Chi-square statistic among algorithms and combination of parameters applied. Marascuilo test results on the comparison of combination of parameters (MM + GL, MM + H, MM + GL + H) per algorithm (J48, Random Tree, REPTree, LMT and Majority Voting) and on the comparison of algorithms per combination of parameters applied. Table S3. Confusion Matrices generated by Weka 3.8.3 software showing the false positives, false negatives and true negatives resulted from each algorithm (J48, REPTree, Random Tree, LMT and Majority Voting) of each combination of parameters (MM + GL + H, MM + GL, MM + H and MM).