Merging of texts

Merging of texts/databases

When a user or an information specialist searches databases, he/she has access to a lot of different documents, document representations and subject access points. Each of the possible access points is formed by some human agent (or by a machine programmed by a human agent). Every element has its own history, and it has been formed by some implicit or explicit goals and theories.

For example, an author writes an article in a given journal in a given discipline. The way he expresses himself is intended for the reader of that specific journal. That journal may be indexed in subject bibliography and thus information from the specific journal is merged with many other journals. Again it may be indexed in databases such as social sciences citation index and merged with journal from many disciplines. Again a searcher may do a one-search in which all citation databases are merged.

A given database can be a merging of what were once different separate databases. In the original databases the access points were perhaps not explicit about some aspects of the subject matter because this was implicit in the delimitation of the database. For example, if you search for "lead" in PsycINFO, there is no need to indicate that you are searching the effects of lead on behavior: it is implicit in PsycINFO that all records are on animal or human psychology/behavior. However, if the records in PsycINFO are merged with the records in Chemical Abstracts, you will have to change your search strategy and specify that you are searching studies on how lead influences behavior and performance. This new strategy would probably be less than optimal regarding the part of the records originally indexed in PsycINFO (because implicit information is lost by the merging).

At another level PsycINFO can be seen as a merging of records which were once presented in individual journals, some of which may be American, some European, some behaviorist, other psychoanalytic, etc. Originally, to the readers of those journals their selection policy and their way of writing titles and composing articles reflected some implicit meanings in those journals. By making a controlled vocabulary, a classification scheme, a certain structure in the records and so on the people behind PsycINFO made certain decisions which were colored by their view of knowledge. For example, Roberts (1985) showed that most thesauri in social sciences were inspired by natural sciences, and were probably designed according to the principles, which were more suitable for documents belonging to natural sciences than for social sciences for which they were intended. Such (more or less implicit) theories of knowledge can be in harmony or conflict with the (more or less implicit) views of knowledge represented by the indexed journals. Both views can again be more or less in harmony or conflict with the implicit or explicit view of knowledge in the query, which again can be more or less in harmony or in conflict with the user's real information need. According to modern semiotic theories also the single document should be understood as a merging of several texts. This is called "intertextuality".” (Hjørland, 1998, p. 27).

Literature:

Buckland, M.; Jiang, H.; Kim, Y. & Petras, V. (2001). Domain based indexes: Indexing for communities of users. In: 3e Congrès du Chapitre français de L'ISKO, 506 juillet 2001. Filtrage et résumé informatique de l'information sur les réseaux. Paris: Université Nanterre Paris X. pp 181-185. http://metadata.sims.berkeley.edu/papers/ISKObuck.pdf

(Visited June 27, 2004).

Gazan, R. (2003). Metadata as realm of translation: Merging knowledge domains in the design of an oceanographic information resource. Knowledge Organization,30(3-4), 182-190.

Hjørland, B. (1998). Information retrieval, text composition, and semantics. Knowledge Organization, 25(1/2), 16-31. Available at: http://www.db.dk/bh/publikationer/Filer/ir_semant_2.pdf

Maniez, J. (1997). Database merging and compatibility of indexing languages. Knowledge Organization, 24(4), 213-224.

Roberts, N. (1985). Concepts, structures and retrieval in the social sciences up to c. 1970. Social Science Information Studies, 5, 5567.