Machine learning recovers folk classification of Banisteriopsis caapi from herbarium leaves an ayahuasca liana

Machine learning analysis of morphological traits within a single plant species shows only partial agreement with traditional ethnobotanical classifications. Confusion matrix and similarity network analyses revealed that automated methods can validate folk taxonomies by detecting subtle variation, though they do not fully replicate previous classification systems. Integrating indigenous knowledge with computational approaches enables systematic assessment of how local communities categorize biological diversity.

Study at a glance

Abstract

reflects morphological overlap. Confusion matrix and similarity network analyses showed only partial agreement with previous ethnobotanical classifications. Focusing on subtle variation within a single species, this study demonstrates that integrating traditional knowledge with machine learning enables automated validation of folk taxonomies.