Project Name:  UMLS Metathesaurus

Project URL: http://www.nlm.nih.gov/pubs/factsheets/umlsmeta.html

Project description:  The UMLS consists of three Knowledge Sources: the UMLS Metathesaurus, the SPECIALIST lexicon, and the UMLS Semantic Network. The Metathesaurus is a database containing semantic information about biomedical concepts, their various names, and the relationships among them.

The Metathesaurus is built from over 100 biomedical source vocabularies, some in multiple languages. The 2003 edition includes 875,255 concepts and 2.14 million concept names. The UMLS Semantic Network is used for mapping index terms from different thesauri through its 134 semantic types which provides a consistent categorization of all concepts represented in the Metathesaurus.

 

Provisional Checklist

Please use the following form to evaluate semantic interoperability projects. 

1. Types of data being integrated

                             Does the project have:

    (a) different controlled vocabularies in same language?

    (b) different controlled vocabularies in different languages?

    (c) different classification schemas (e.g., DDC, UDC, LCC)?

             If yes, which ones?      

    (d) controlled vocabularies combined with classification schemas?

    (e) different metadata framework schemas (e.g., XML, MARC, Dublin Core)?  

             If yes, which ones? Dublin Core

   (f) different communication protocols?

    (g) other:      

 

2. Autonomy and Integrity of Constituent Parts

    (a) Is standardization, reconciliation, or conversion of semantic data                       reversible? 

        (a.1) Can precoördinated strings, once filtered or deconstructed for semantic matching, later be put back together again?

    (b) Is full complement of metadata and indigenous subject hierarchies preserved?

            If so, how? Through linking

    (c) Does project rely on principle of least common denominator?

If so, many data sets may be able to coexist in database, but given resulting stripped-down or ‘dumbed-down’ resource descriptions, the database may no longer serve the interests of readers. (cf. recently cited problems with Dublin Core20   How does the use of least common denominator effect the quality of service?     

            (d) How is data stored: gathered into a union catalog (e.g., American Memory Project, NSDL), vs. distributed database?  Distributed database

            (e) How are metadata (including SI links) stored?  (e.g., via authority records, concordance tables, a central switching language, semantic networks, lexical databases, semantic layers, etc.)  Linking records

 

 

 

3. Reconciliation of heterogeneous vocabularies

            (a) How is correlations established when a single term in one source has no equivalent term in the other?  Linking 

            (b) Certain vocabularies are highly structured and hierarchical, while others contain terms lacking any structure at all aside from serial numbers or other unique identifiers. How are these differences reconciled?   By adding certain basic information to each concept and establishing new relationships between terms from different source vocabularies

            (c) How are conflicts resolved when an established heading in one vocabulary matches a cross reference in other vocabularies? (E.g., Tumors is an established LCSH heading, but in MeSH it is a cross reference to Neoplasms; and vice versa).   By linking concepts that are similar along some dimension

 

           (d) If multiple vocabularies are used in a single bibliographic record, and the headings from such vocabularies are identical (after normalization), how are duplicate retrievals handled?   Through linking

 

 

4. Effective and Efficient Resource Discovery (Precision and Recall), Satisfying User Needs

     (a) Does project provide high or satisfactory levels of precision and recall?

 

Precoordination  

           (b) To what extent does project rely on precoördination?

 

        If mostly post-coordinate, then:

i)                    by what means is recall maximized?    

 

ii)                   by what means is precision maximized?     

 

 

   (d) Does project provide faceted approach (facilitating polysemy) while retaining option for browsable hierarchy (facilitating navigation)?

          (e) Are the following objectives and functions supported in the S.I. environment?

        i)         Locate entities in the system via surrogates (find)

         ii)      Identify a surrogate that matches an entity (collocate)

         iii)     Select an entity appropriate to a user’s need via surrogates (choice facilitation)

         iv)       Obtain access to the entity via the system and its surrogates (acquisition)

        v)      Navigate the system and its surrogates (navigation)

   (f) Has developer released beta version for general testing?

   (g) Have user satisfaction surveys been conducted?

 

 

5. Ease of Use (this is actually part of our definition, i.e., SI should function “without special effort by the user,” (where “users” include information creators and managers, and end-users)).

    (a) Intuitive interface for data entry, searching, browsing, etc.?

    (b) Automate validation, mapping, metadata extraction, etc., as much as possible?

    (c) Availability of documentation?

 

6. Long-term viability

    (a) Master plan for life-cycle management and data migration?

    (b) Reliance on open-source international standards versus proprietary standards?

  (c) viable business model (e.g., not based exclusively on research grant with likely expiration)?