This module defines two Bayesian networks:
- A network based on the Linnaean classification system. This network is based around the kingdom - phylum - class - order - family - genus - species hierarchy used in biological scientific nomenclature. It uses higher-order information in the supplied template to confirm and disambuguate evidence. It includes a number of other elements that prove useful when dealing with taxa. Also present here is a classification matcher that tests for a number of conditions, such as parent-child synonyms, multiple synonyms and the like and produces a preferred match.
- A network for vernacular names. This network uses ancially information, such as sex, location or lifeStage to help disambiguate usaes of a name.
Each network has an analyser that (particularly in the case of the Linnaean network) attempts to beat supplied information into a common form.
Note that to correctly build, this module needs to have the pre-built name indexes installed. See the parent README for more information. If you do not have them and need to build them, skip tests so that you can run the ALA index builder.