Classifying minerals and their related names in a relational database

Abstract

The categorisation of minerals and their related names, such as synonyms, obsolete or historical names, varieties or mixtures, is an asset for designing an interoperable and consistent mineralogical data warehouse. An enormous amount of this data, provided by mindat.org and other resources, was reviewed and analysed during the research. The analysis indicates the existence of several categories of (1) the abstract titles or designations representing the link to the original material or a group of names or substances without actual physical representation, and (2) the unique names representing actual physical material, compounds, or an aggregate of one or more minerals. A revision of the dependency between the categories attributes stored in a database (e.g. chemical properties, physical properties) and their classification status assigned allowed us to design a robust prototype for maintaining database integrity and consistency. The proposed scheme allows standardisation and structuring of officially regulated and maintained species, e.g. IMA-approved, and, in addition, unregulated ones.