2010 TestSuiteDesignForBioOnt

From GM-RKB
Jump to navigation Jump to search

Subject Headings:

Notes

Quotes

Abstract

  • Systems that locate mentions of concepts from ontologies in free text are knownas ontology concept recognition systems. This paper describes an approach tothe evaluation of the workings of ontology concept recognition systems throughuse of a structured test suite and presents a publicly available test suite forthis purpose. It is built using the principles of descriptive linguisticfieldwork and of software testing. More broadly, we also seek to investigatewhat general principles might inform the construction of such test suites. Thetest suite was found to be effective in identifying performance errors in anontology concept recognition system. The system could not recognize 2.1% ofall canonical forms and no non-canonical forms at all. Regarding the questionof general principles of test suite construction, we compared this test suiteto a named entity recognition test suite constructor. We found that they hadtwenty features in total and that seven were shared between the two models,suggesting that there is a core of feature types that may be applicable to testsuite construction for any similar type of application.

,

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2010 TestSuiteDesignForBioOntK. Bretonnel Cohen
William A. Baumgartner Jr
Christophe Roeder
Lawrence E. Hunter
Karin Verspoor
Test Suite Design for Biomedical Ontology Concept Recognition Systems