README.format.mm.pod
This file briefly describes the .mm format used by the CuiTools package.
The .mm format is the xml-like format that is used by the CuiTools package.
The complete context is displayed in the <context line="<context>"/> tags.
The sentences in the context are broken up and displayed in the <sentence tw="<target word, if exists in the sentence>" instance="<id>" line="<sentence>" tags. There may be more than one sentence per context and the sentence id may or may not be available.
The sentences are broken up and displayed in the <phrase tw="<target word, if exists in the phrase>" hw="<head word of the phrase>" line="<phrase>"/> tags. There is more than one phrase associated with a single sentence. The "tw" contain the target word if it exists in the phrase.
The phrases are then broken up into their individual words using the <token word="<word>" pos="<part of speech>" tags. If the word is the target word then instead of "token" the tag will display "target". Inside the token tags there may exist a mapping tag which contain information from the National Library of Medicine's MetaMap Transfer (MMTx) program. This information includes the concept unique identifiers (CUIs) associated with that token. A mapping tag would look as such:
<mapping rank="<rank>" score="<score>" umls_cui="<the UMLS concept unique identifier>" umls_concept="<UMLS concept>" semantic_types="<semantic types>"/>
The CUIs associated with a token are assigned a score and are ranked based on this score by MMTx. The mapping tag includes the rank and the score information. The UMLS concept unique identifier is an ID, such as "C0003826", indicating the the UMLS concept. The UMLS concept is the written form of the CUIs. The UMLS concept for cui "C0003826" is "Art". The semantic types are the types that are associated with that cui such as "ocdi". There may exist more than one semantic types, in which case they are comma separated. A complete listing of semantic types can be found: http://mmtx.nlm.nih.gov/semanticTypes.shtml.
There may or may not exist a mapping tag for each token/target tag.
Example of .mm format for the context line: "Paul was name Art magazine's top collector".
<corpus lang='en'>
<lexelt item="art" senses="art">
<instance id="art.30002" alias="art">
<answer instance="art.30002" senseid="art"/>
<context line="Paul was name Art magazine's top collector"/>
<sentence tw="art" id="art.30002" line="Paul was name Art magazine's top collector">
<phrase tw="" head="" line="Paul">
<token word="paul" pos="noun">
</token>
</phrase>
<phrase tw="" head="" line="was">
<token word="was" pos="aux">
</token>
</phrase>
<phrase tw="" head="name" line="name">
<token word="name" pos="verb">
<mapping rank="1" score="1000" umls_cui="C0027365" umls_concept="Name" semantic_types="idcn,inpr"/>
<mapping rank="2" score="966" umls_cui="C0233735" umls_concept="Naming" semantic_types="menp"/>
</token>
</phrase>
<phrase tw="art" head="magazine" line="Art magazine">
<target word="art" pos="noun">
<mapping rank="1" score="694" umls_cui="C0003826" umls_concept="Arts" semantic_types="ocdi"/>
<mapping rank="2" score="694" umls_cui="C0220786" umls_concept="Products of the Arts" semantic_types="mnob"/>
<mapping rank="3" score="694" umls_cui="C1140123" umls_concept="WHO Adverse Reaction Terminology" semantic_types="inpr"/>
</target>
<token word="magazine" pos="noun">
<mapping rank="1,2,3" score="861" umls_cui="C0162443" umls_concept="Magazines" semantic_types="inpr,mnob"/>
</token>
</phrase>
<phrase tw="" head="" line="''s">
<token word=" s" pos="aux">
</token>
</phrase>
<phrase tw="" head="collector" line="top collector">
<token word="top" pos="noun">
<mapping rank="1" score="694" umls_cui="C1552827" umls_concept="Table Cell Vertical Align - top" semantic_types="spco"/>
<mapping rank="2" score="694" umls_cui="C0000811" umls_concept="Termination of pregnancy" semantic_types="topp"/>
<mapping rank="3" score="661" umls_cui="C0076833" umls_concept="container tops" semantic_types="medd"/>
</token>
<token word="collector" pos="noun">
<mapping rank="1,2,3" score="827" umls_cui="C0180011" umls_concept="Collectors" semantic_types="medd"/>
</token>
</phrase>
</sentence>
</instance>
</lexelt>
</corpus>
Bridget McInnes, University of Minnesota, Twin Cities
Copyright (c) 2007-2011,
Ted Pedersen, University of Minnesota, Duluth. tpederse at umn.edu
Bridget McInnes, University of Minnesota, Twin Cities bthomson at cs.umn.edu
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, write to
The Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.