Supplementary information: http://www.nature.com/nature/journal/v438/n7070/extref/438900a-s1.doc
Evaluation in Knowledge Organization
Different approaches to knowledge organization tend to have different ideals on how best to evaluate systems and processes.
Systems are evaluated on the basis of how well they reflect literary warrant as related to current scientific standards. Subject experts used to construe/revise/evaluate systems. (Some resemblance with peer-review). |
|
1a. Business- or management like approaches | Systems are evaluated on the basis of management criteria: What are the costs and benefits from a management perspective. This perspective also emphasis standardization in KO |
Systems are evaluated on the basis of their logical principles and how consistent those principles have been applied. |
|
Systems are evaluated on the basis of experiments, e.g. the measuring of recall and precision. |
|
Systems are evaluated on the basis of user preferences or on the basis of cognitive models. |
|
(The same as Information retrieval tradition?) |
|
Systems are evaluated on the basis of literary warrant, how they reflect domain specific knowledge emphasizing how they relate to different views and epistemologies in the knowledge represented. Also pragmatic analyses concerning what goals, interests and consequences are involved. |
Wikipedia has been compared with Encyclopedia Britannica in an article in Nature. See Giles (2005), Encyclopedia Britannica (2006) and Nature's response March 23, 2006. The method used for comparing those two knowledge organization systems is the peer-review system known form scientific journals. The reviewers were asked to look for three types of inaccuracy: factual errors, critical omissions and misleading statements. Such criteria are somewhat different from criteria, which should be used for evaluating semantic tools such as classification systems and thesauri. In those cases are the most important dimensions 1) the coverage (are all important concepts and terms included or are there systematic biases? 2) How are the terms defined? 3) Which semantic relations are provided?
Literature:
Encyclopedia Britannica (2006). Fatally Flawed. Refuting the recent study on encyclopedic accuracy by the journal Nature. http://corporate.britannica.com/britannica_nature_response.pdf. Nature's response March 23, 2006: http://www.nature.com/press_releases/Britannica_response.pdf
Giles, J. (2005). Special Report: Internet encyclopaedias go head to head. Nature, 438, 900-901. Available: http://www.nature.com/nature/journal/v438/n7070/full/438900a.html
See also: Critique in Knowledge Organization; Errors in Knowledge Organization; Information retrieval evaluation
Birger Hjørland
Last edited: 02-04-2006