NAME
Todo List for UMLS-Similarity
SYNOPSIS
Plans for future versions of UMLS-Similarity
TO DO LIST
1. Add configuration information to the cache files.
2. Add option such that after the shortest path between the two concepts
is of a certain threshold just return a similarity score of zero. There
is something similar to this in WordNet:: Similarity. I think this would
be very useful when using the --realtime option
3. Explore using measures of association rather than frequency in the
vectors for the vector measure
4. Explore linking WordNet and the UMLS together outside of the --dict
file option for the relatedness measures
5. Add additional defaults for the icpropgation and vector matrix files.
Right now vector matrix comes from clinical records, I think we should
explore using the bigrams over the entire Medline provided by NLM. The
bigram files is large and therefore thresholds here become really
important otherwise we get killed on time.
AUTHORS
Bridget T. McInnes <bthomson at cs.umn.edu>
SEE ALSO
COPYRIGHT
Copyright (C) 2007-2011 Ted Pedersen, Bridget T. McInnes, Ying Liu,
Serguei Pakhomov
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.2 or
any later version published by the Free Software Foundation; with no
Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts.
Note: a copy of the GNU Free Documentation License is available on the
web at <http://www.gnu.org/copyleft/fdl.html> and is included in this
distribution as FDL.txt.