UMLS::Similarity

This is a Perl module that implements a variety of semantic similarity and relatedness measures based on ontologies and terminologies found in the Unified Medical Language System (UMLS). The measures assign numeric values between pairs of medical concepts indicating how similar or related they are.

We have a mailing list designed to support users of UMLS::Similarity.

We presented a tutorial at ACM IHI 2012 that contains extensive material about UMLS::Similarity and the underlying ideas. See our slides!

Local Installation

The current version of UMLS::Similarity is always available at CPAN.
If you install locally, you'll need to install UMLS::Interface and various other supporting software (like mysql and the UMLS) first. These instructions should help.
Once you have UMLS::Similarity installed, you may want to start off by running umls-similarity.pl from the command line.

Web Services

Try the UMLS::Similarity Web Interface here or here. This does not require any installation on your system, so is a good way to get started.

Documentation

Both UMLS::Similarity and UMLS::Interface contain extensive documentation.
See the history of UMLS::Similarity.
If you use UMLS::Similarity or UMLS::Interface, please cite the following paper:
- UMLS-Interface and UMLS-Similarity : Open Source Software for Measuring Paths and Semantic Similarity (McInnes, Pedersen, and Pakhomov) - Appears in the Proceedings of the Annual Symposium of the American Medical Informatics Association, Nov 14-18, 2009, pp. 431-435, San Francisco, CA
We are using these tools as a part of our larger efforts in Health Informatics and Pharmacy Informatics.

Data

The reference standards we've used to evaluate UMLS::Similarity are available here.

Acknowledgments

The development of UMLS::Similarity and UMLS::Interface has been supported by an RO1 grant from the National Institutes of Health, National Library of Medicine (#1R01LM009623-01A2).

By: Ted Pedersen - tpederse AT d umn edu