Subject and content categorization of human nutrition documents in agricultural and medical databases
1997
Bartol, T. (Ljubljana Univ. (Slovenia). Biotechnical Fac., Slovenian National AGRIS Centre)
Our purpose was to assess the presence of human nutrition-related documents (d) in selected large agricultural (AGRICOLA, AGRIS, CAB) and medical (MEDLINE) databases (db) and especially to thereby illustrate broad non-descriptor classification methods. We based our searches on Categories (C) and Category Codes. The three agricultural db provide fairly similar classification systems with well defined human nutrition (S) and also food processing (Q) categories. Documents are rarely indexed with both C together what is unusual, given such Q C as Food Composition and Quality which are clearly also S-related. It might be inferred that the indexers are somehow biased towards processing and technology aspects. Searches for S should therefore not exclude Q C. It is also possible to define the two C in MEDLINE, however, due to the medical nature of this db any such C can strongly interact with clinical aspects, so it is much more difficult to define "clean" S and Q C, which, in this case, yield lower search results. Overall yearly document growth (1993) of all d is in the agricultural db the highest in AGRIS, whereas Q, S, (Q not S), (S not Q) and (Q or S) search frequencies are the highest in CABI, where, however, an everage d is assigned notably more Category Codes, what might result in seemingly greater retrieval of Q or S related d in CABI. Specialized db for food and nutrition (FSTA, Food & Human Nutrition, Food Intelligence) were not used in our research as most of their records are Q and S related.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by University of Ljubljana