Fig 1.
Centripetal and Discoid débitage rhythm, directions and main blanks types.
A) Modified from [37].
Table 1.
Predetermination criteria for Discoid and Levallois Methods Centripetal Recurrent.
Table 2.
Discoid/Centripetal Levallois’ experimental lithic inventory.
Fig 2.
Flake variables measured for analyses.
Fig 3.
Flake variables measured for symmetry index.
Table 3.
Details on the performance of the seven Machine Learning algorithms compared.
Table 4.
Details on the performance of the seven Machine Learning algorithms compared after hyperparameter optimization.
Fig 4.
Importance of variables for the first tuned model of Random Forest.
Here, “MeanDecreasGini” measures the total decrease in node impurity at each split, weighted by the proportion of samples reaching that node in each individual tree. The more the Gini Index decreases for a feature, the more important it is.
Fig 5.
Relationship between Out-of-bag error and optimal mtry value for Random Forest.
Fig 6.
Importance of variables for the final Random Forest model, without including “technology classification” variable.
Fig 7.
Error variation for both the two classes based on the number of trees.
Table 5.
Details on the performance of the seven Machine Learning algorithms compared after hyperparameter optimization and without “technology classification” variable.