The first part of the abbreviation identifies the type of practitioner:Rx radiologist,IT IT professional; the second part of the abbreviation identifies the participant's country:B Belgium,D Germany,Fr France,I Italy,IL Israel,P Poland,TK Turkey,UK United Kingdom; the third part of the abbreviations is a number to distinguish different representatives of the same country.
> We tested the three abbreviation identifying modules on the Ab3P, BIOADI, MEDSTRACT and Schwartz and Hearst corpora as shown in Table 3. Results are based on the new gold standard annotations in the four abbreviation corpora.
- Each part of an abbreviation
The meaning of abbreviations that were identified on this basis was looked up in the documents or the Internet in order to assign categories to various abbreviations such as RROP (Regionales Raumordnungsprogramm; regional spatial planning program), SWE (Stadtwerke Emden; public utilities of Emden), or LED (light emitting diode).
For abbreviations of genes identified by SuperSAGE, see Additional file 3 or [ 89].
We identified abbreviation detection as a supportive task for sentence delineation.
One important enhancement to these corpora encouraged by the BioC format was the specification of the exact location of each identified abbreviation definition.
Regarding abbreviation resolution, we adapted a simple but effective abbreviation definition recognizer [28], which is based on a set of pattern-matching rules to identify abbreviations and their full forms.
We used an adapted version of the Hearst and Schwartz algorithm [11] to identify abbreviations of entities found by our system.
However, we did not implement or tackle the problem of identifying abbreviations because we believe it is a different problem that can be solved with existing third party solutions such as [16].
In addition, if a system fails to identify abbreviations, their interpretation by mapping to full forms is impaired.
