Truth table classification and identification
Abstract
A logical basis for classification is that elements grouped together and higher categories of elements should have a high degree of similarity with the provision that all groups and categories be disjoint to some degree. A methodology has been developed for constructingclassifications automatically that gives nearly instantaneous correlations of character patterns of orgnisms with time and clusters with apparent similarity. This means that automatic numericalidentification will always construct schemes from which disjoint answers can be obtained if test sensitivities for characters are correct. Unidentified organisms are recycled through continuous classification with reconstruction of identification schemes. This process is cyclic and self-correcting. The method also accumulates and analyzes data which updates and presents a more accurate biological picture.
- Publication:
-
Space Life Sciences
- Pub Date:
- December 1971
- DOI:
- 10.1007/BF00927988
- Bibcode:
- 1971SLSci...3..135R