Hi S.H., cTAKES does not query the UMLS online version. The cTAKES dictionary is built from a subset of the UMLS Metathesaurus. There are two tools in the cTAKES sandbox, dictionary-gui and dictionarytool that you can use to build a dictionary from the subset of the UMLS Metathesaurus that you installed. (they use the META files, though - not the MySQL database) The tools allow you to choose what TUIs, vocabularies, etc. to include in your dictionary. These choices affect the dictionary's size.
I hope that helps, Jessica On Tue, Oct 4, 2016 at 12:36 PM, SH.Chou <[email protected]> wrote: > Hi All, > I just started to use cTAKES, and have a question regarding the data > size of UMLS 2011ab (the default dataset in cTAKES) and new 2016aa. > I install 2016aa in MySQL database, the data size is about 14G~, but the > 2011ab in cTAKES is just 2G~. I wondered if cTAKES use UMLS API and submit > words to query UMLS online version? > Or cTAKES compressed 2011ab (using HSQL?). > > Thanks, > S.H. > > >
