Missouri Botanical Garden Open Conference Systems, TDWG 2016 ANNUAL CONFERENCE

Font Size: 
Ontology-driven taxonomic workflows for Afrotropical Bees
Aurona Gerber

Last modified: 2016-09-29


This poster presents the results of an investigation into the use ontology technologies to support taxonomy workflows. Taxonomy in biology is often referred to as the science of naming and grouping biological organisms into a hierarchy. A core function of biological taxonomy is thus the classification and revised classification of biological organisms into an agreed upon taxonomic structure such as the taxonomy of Linnaeus based on sets of shared characteristics.

Taxonomy is considered to be under pressure due to high demand for identification services, as well as the lack of expertise and resources due to the highly specialised and time-consuming functions taxonomists perform. This research investigated how computer technologies can assist and support biological taxonomy and taxonomists, first in describing species and their relationships to one another, and second, in the identification of specimens.

Recent developments in knowledge representation within Computer Science include the establishment of computational ontologies as a knowledge representation technology particularly suitable to support classification functions such as those exhibited in biological taxonomy. Computational ontologies are knowledge representation mechanisms based on description logics (DLs), which are decidable fragments of first-order logic. This logic base of computational ontologies provides several advantages for the computerized capturing and manipulation of knowledge, and has resulted in the specification of OWL, the Web Ontology Language, as a W3C standard for expressing meaning and semantics in a computer readable format. The set-theoretical basis of computational ontologies is the aspect that makes them particularly suitable for classification functionality such as required in biological taxonomy.

Using a specific genus of Afrotropical bees, this research had as its goal the capturing and representation of the taxonomic knowledge base into an OWL ontology, exploring the use of available reasoning algorithms to draw inferences that support the necessary taxonomy workflows and implementing a Web-based application encapsulating the components for users. The application uses the ontology and demonstrates some derived taxonomic functions namely: the identification (keys), as well as the description and comparison of taxa (taxonomic revision). The contribution includes an approach towards representing taxonomic knowledge in an ontology, presenting a reusable and standardised computable knowledge base for the taxonomy of Afrotropical bees (the ontology), adapted reasoning functionality to assist with taxonomic workflows over the ontology, as well as a Web-based application and evaluation thereof.