Building a deep ontology-based herbal medicinal plant search system

Traditional medicine comprises practical experience accumulated over thousands of years. Despite the abundance of published works on medicinal plants, it remains relatively complex for general readers to learn how to use them. This paper presents DOMLS, a deep ontology-based herbal medicinal plant search system equipped with a mechanism for analyzing both text and image queries. The system allows searching for medicinal plants based on various criteria, including images, names, functions, diseases, taboos, and valuable parts. We create an ontology for text search and extend the search result, and use the YOLOv5 deep learning model for image classification. We test the system with a dataset collected from medicinal plants in health centers and medical gardens to demonstrate its high accuracy.