JMIR Publications

ABSTRACT

Background:

Cross-mapping establishes an equivalence between terms from different terminology systems. The number of terms to be mapped determines an extensive, tedious, thorough work that is susceptible to errors that can be minimized by self-matching, which is a mapping process that uses computational tools.

Objective:

To compare the results of term mapping processes, both manual and automated.

Methods:

A descriptive, quantitative study using the results of two mapping processes as an empirical basis: manual, which used 2,638 nursing terms from a Brazilian university hospital and the International Classification for the Nursing Practice (ICNP); and automated, which used the same terms as the university hospital and the primitive terms of the ICNP, through an algorithm based on rules of natural language processing called MapICNP. A comparison was performed via equality and exclusivity assessments of new terms of the automated process and of candidate terms.

Results:

The self-combining process mapped 21.56% of the source bank’s terms as identical, and manual process, 24.82%. Regarding the new terms, the automated process mapped 39.08% of the source bank’s terms, while the manual mapped 34.79%. In particular, manual mapping identified 101 terms as identical and 429 as new, whereas automated mapping identified 20 identical terms and 209 new. It was possible to establish an equivalence between 48 terms of the source bank with those of ICNP; 100 candidate terms had a semantic relationship with the source term.

Conclusions:

The automated and manual processes map identical and new terms in a similar way and can be considered complementary. Direct identification of identical terms and the offer of candidate terms through the automated process facilitate and enhance the results of the mapping, whose confirmation of precision requires analysis by the researcher.