Machine learning recovers folk classification of Banisteriopsis caapi from herbarium leaves an ayahuasca liana
Scheila Cristina Biazatti, Deborah Bambil, Rômulo Môra, Lúcio Flávio de Alencar Figueiredo, Regina Célia de Oliveira
iScience April 15, 2026 Peer reviewed DOI: 10.1016/j.isci.2026.115753 via OpenAlex
Summary
Integrating traditional knowledge with machine learning allows for automated validation of folk taxonomies, revealing that there is only partial agreement with previous ethnobotanical classifications. The study highlights the importance of subtle variations within a single species and reflects morphological overlap among them.
Study at a glance
| Key finding | The study demonstrates that combining traditional knowledge with machine learning can validate folk taxonomies, showing only partial agreement with prior classifications. |
|---|
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.