We had similar discussion about including the cue words for negation or uncertainty, etc. How would you use these cue words/dictionary words after you've already done your NER or attribute discovery?
I think this is a little too similar to either using cue-word-based rules for everything or basically including whole feature vectors for each decision made. stephen On 8/21/13 1:26 PM, "Masanz, James J." <[email protected]> wrote: >other nice-to-have along these lines would be >(1) the words within the text that caused the annotation to be created. >(2) the words of the dictionary entry that caused the annotation to be >created. > >(1) is especially useful when there are intervening words within the >text. >For the phrase "cancer of colon, lung and liver" annotated with 93870000 >due to the words "cancer" and "liver" > >(2) is especially useful when the preferred term does not share any words >in common with a dictionary entry (due to synonyms) > >-----Original Message----- >From: [email protected] >[mailto:[email protected]] >On Behalf Of Pei Chen (JIRA) >Sent: Wednesday, August 21, 2013 1:08 PM >To: [email protected] >Subject: [jira] [Created] (CTAKES-226) CEM Templates- In the xxxMentions >locations/severity store Annotation instead of the BinaryTextRelation > >Pei Chen created CTAKES-226: >------------------------------- > > Summary: CEM Templates- In the xxxMentions >locations/severity store Annotation instead of the BinaryTextRelation > Key: CTAKES-226 > URL: https://issues.apache.org/jira/browse/CTAKES-226 > Project: cTAKES > Issue Type: Improvement > Affects Versions: 3.1 > Reporter: Pei Chen > Fix For: future enhancement > > >> Hi James/Steven, >> In the common type system/template fillers, do you recall why we >> stored the TextRelation instead of the resolved annotation? >> >> For example, in SignSymptomMention, getBodyLocation() returns >> LocationOfTextRelation. >> So in order to actually get the AnatomicalSiteMention, you would have >> to look inside LocationOfTextRelation arg1 or arg2. >> >> I think it be more intuitive and simpler for consumers of the CEM's to >> just store the AnatomicalSiteMention? Is there a use case I am >> missing where someone would want something different than the >> SignSymptomMention.getBodyLocation() other than the actual >> AnatomicalSiteMention? > >Email Thread: >http://mail-archives.apache.org/mod_mbox/ctakes-dev/201308.mbox/%3C924DE05 >C19409B438EB81DE683A942D910588E15%40CHEXMBX1A.CHBOSTON.ORG%3E > >-- >This message is automatically generated by JIRA. >If you think it was sent incorrectly, please contact your JIRA >administrators >For more information on JIRA, see: http://www.atlassian.com/software/jira
