A Proposed System to Identify and Extract Abbreviation Definitions in Spanish Biomedical Texts for the Biomedical Abbreviation Recognition and Resolution (BARR) 2017

Biomedical Abbreviation Recognition and Resolution (BARR) is an evaluation track of the 2nd Human Language Technologies for Iberian languages (IberEval) workshop, which is a workshop series organized by the Sociedad Española del Procesamiento del Lenguaje Natural (SEPLN). In this first edition of BARR, the focus is on the discovery of biomedical entities and abbreviation, and relating detected abbreviations with their long forms. This paper describes the system and the approach presented in this track. We develop a ruled-based system using an adapted version of the algorithm for extraction of abbreviations and their definitions from biomedical text proposed by Schwartz & Hearst.