The document discusses perspectives on metadata from web resources and database systems. It describes how metadata comes in many forms and serves various purposes, such as supporting discovery and identification of information resources on the web (resource metadata), and ensuring consistency and analysis of structured data in databases (metadata in database systems). Resource metadata commonly follows standards and is stored separately from the resources it describes, while database metadata includes both structural metadata describing data organization and content metadata in the form of data dictionaries.

1. Matthew
Brush
Ontology
Development
Group
OHSU Library,
DMICE
METADATA
PERSPECTIVES
FROM THE WEB
AND DATABASES
SYSTEMS
Sept 17, 2014
brushm@ohsu.edu

2.  “Data about Data”
 “Data” broadly covers any information resource
 digital or physical
 narrative, multimedia, structured
 raw data, processed data, aggregates of datasets, or
discrete elements within data sets
 More formally, “Metadata is structured information
that describes, explains, locates, or otherwise makes
it easier to retrieve, use, or manage an information
resource”
METADATA
(NISO (2004) Understanding Metadata. Bethesda, NISO Press )

3.  Descriptive metadata: supports discovery and identification
 e.g. title, author, identifiers, subjects, keywords
 Structural metadata: describes how the components of a
resource are organized
 e.g. table of contents for a book, schema of database tables,
manifest of files in an aggregate ‘research object’
 Administrative metadata: helps manage the resource
 Technical - describes technical aspects of a resource
 e.g. file type, version information, how/when created
 Rights management - explains intellectual property rights
 e.g. licensing, use restrictions, privacy concerns
 Preservation - supports maintenance and archiving of a resource
 e.g. provenance/ownership, history of use, authenticity
METADATA SERVES MANY PURPOSES . . .
http://www.niso.org/publications/press/UnderstandingMetadata.pdf

4. Metadata comes in many forms, serves many needs,
and operates in very diverse settings
 I. Resource metadata (on the web)
 Target: information resources as a whole
 1o
Goals: resource discovery and use
 Form: structured, separate records
 Users: everyone
 Standards: many metadata frameworks/vocabularies
 II. Metadata in database systems
 Target: structured data and data elements
 1o Goals: data consistency, aggregation, analysis
 Form: ER diagrams, summary tables, data dictionaries
 Users: professional data administrators and scientists
 Standards: metadata and CDE registries
. . . AND OPERATES IN MANY CONTEXTS

5. I. Resource Metadata (on the web)
A. Overview
B. Examples
C. Metadata Frameworks
i. Schema
ii. Vocabularies
iii. Conceptual Models
iv. Practical Specifications
v. Encoding Specifications
D. Metadata Storage and Retrieval
II. Metadata in Databases Systems
A. Overview
B. Data Elements
C. Data Dictionaries
D. Common Data Elements (CDEs)
E. CDE Registries
OUTLINE

6.  Metadata in the world that all of us have used and
created in work and life
 Attached to information resources we find on the web
 books, videos, images, websites, datasets, . . .
 Helps us to find a resource and understand what it is
and how to use it
I. RESOURCE METADATA (ON THE WEB)

7. Descriptive
Structural
Administrative
Book Catalog Record
http://ohsulibrary.worldcat.org/title/metadata/oclc/225088362

8. Descriptive
Structural
Administrative
Digital Photograph Library
http://crdl.usg.edu/cgi/crdl?query=id:highlander_highlanderphotos_p2-wi3-3

9. Data Set Description
http://datadryad.org/resource/doi:10.5061/dryad.4ms68
Research Data Sets
and Files (datadryad.org)
Data File Description

10.  Resource metadata is increasingly structured
according to established schemas and standards
 Many standards exist that vary in their:
 complexity (schemas, specifications, conceptual models)
 targets (music, video, images, books, art, datasets)
 goals (descriptive, administrative and preservation)
 communities served (libraries, museums, research)
 Benefits:
 leverage existing resources
 vetted by community
 interoperability and integration
STANDARDS ARE KEY

11. Normative standards for metadata are
captured in metadata frameworks.
There are five possible components of a
metadata framework:
A. Schema
B. Vocabularies
C. Conceptual Model
D. Practical Specifications
E. Encoding Specifications
METADATA FRAMEWORKS

12.  Core of any framework – specifies the categories of
information recorded
 Comprised of a set of data elements along with
descriptions of their attributes and rules for use
 attributes described should minimally include an identifier
and/or name and a definition of each element
 Can also specify data types and ‘value domains’ that
describe allowable values for a given element
 e.g. term lists, CVs, ontologies
 Example schema: Dublin Core, LOM, HCLS Dataset Std.
A. METADATA SCHEMA

13.  First effort at standardizing metadata to improve
resource discovery on the web
 Very simple core schema consisting of 15 general
data elements representing properties of a
information resource, with no value restrictions.
 Data Elements: title, identifier, type, description,
creator, contributor, date, subject, format, language,
source, publisher, relation, coverage, rights
 Element Attributes: URI, label, definition, domain,
range, version, comment
EXAMPLE 1: DUBLIN CORE
METADATA INITIATIVE (DCMI)
http://dublincore.org/documents/dcmi-terms/

14.  Extensive set of metadata elements describing
‘learning objects’
 “Any digital or non-digital entity that may be used for learning,
education, or training"
 Based loosely on DCMI schema, but:
 >50 new elements to describe educational attributes of learning
objects
 organizes elements into a hierarchical structure
 provides detailed specifications for allowable values
 supports ‘application profiles’ that extend model for
specific domains
EXAMPLE 2: LEARNING OBJECT
METADATA (IEEE-LOM)
http://www.imsglobal.org/metadata/mdv1p3/imsmd_bestv1p3.html

15. LOM SCHEMA ELEMENTS AND
ATTRIBUTES

16.  The LOM base schema defines 9 categories of metadata elements
 Hierarchical structure supports user understanding, metadata
organization and aggregation for analysis
LOM ELEMENT HIERARCHY

17.  A unified schema that provides all key metadata fields
needed to comprehensively describe research datasets
 what they are, how they are produced, where they are found
 meets pressing need in current research climate to support sharing,
discovery, and re-use of public datasets in a standardized way
 Metadata elements describe general features, identifiers,
provenance and change, availability and distribution, and
dataset statistics
 Comprised entirely of elements (properties) from existing
community vocabularies, e.g. DCMI, DCAT, PROV, VOID,
FOAF
 attributes and rules for element use defined in source schema
EXAMPLE 3: W3C HCLS DATASET
DESCRIPTION STANDARD
http://www.w3.org/2001/sw/hcls/notes/hcls-dataset/

18. B. VOCABULARIES
 Set of terms (often structured) that is used to
constrain entry of metadata values
 Vocabularies represent general concepts
 Word or code lists
 Hierarchical classifications
 taxonomies, thesauri, ontologies
 e.g. ICD9, SNOMed, MeSH, NCIthesaurus
 Authority lists provide controlled names for proper nouns
 FundRef (organizations)
 Global Gazeteer (places)
 ORCID (people)

19.  Open Researcher and Contributor IDentifier (ORCID)
 a nonproprietary alphanumeric code that uniquely identifies scientific
and other academic authors (a persistent “author DOI” for researchers)
 The ORCID identifier set is coming to serve as a de facto
authority list to record persons contributing to scholarly research
products
 ORCIDs facilitates efforts to track productivity, impact, and
attribution based on all scholarly outputs (publications, grants,
datasets, protocols, presentations, abstracts, code, blogs, etc)
 Services can aggregate scholarly outputs for a given researcher
 resolves to a “CV” listing all scholarly contributions linked across various
venues (e.g. Pubmed, Scopus, Slideshare, Figshare, Github, Dryad, . . .)
ORCID AS AN AUTHORITY LIST

20.  An underlying model that describes how all the
information and concepts inherent in a resource are
related to one another
 Metadata Models
 conceptualize the metadata schema itself (hierarchical
relationships or other mappings between elements )
 Domain Models
 conceptualize domain in which the metadata schema
operates (classes of things that are annotated and the
relationships between them)
C. CONCEPTUAL MODELS

21. EXAMPLE METADATA MODEL:
LOM ELEMENT HIERARCHY
The structure of the LOM is an example of a simple conceptual metadata model,
which organizes elements into disjoint hierarchies

22.  The summary level describes the dataset in general
 The version level describes a specific version
 The distribution level describes a representation of a version
EXAMPLE DOMAIN MODEL:
HCLS DATASET ‘LEVELS’
Supports recommendations for how each should be described using the standard

23.  D. Practical specifications for use
 provide guidance for how to apply metadata under a given schema
 e.g. HCLS model provides recommendations when and how to apply
certain elements to types of targets in the domain
 E. Encoding specifications for presentation & exchange
 rules for binding metadata to syntactic formats such as XML or RDF
 e.g. LOM has precise specification for binding to XML or RDF
D/E. SPECIFICATIONS

24. STORING AND ACCESSING
RESOURCE METADATA
 Typically lives separately from annotated resources,
in databases and/or XML files
 Can also be stored within a resource (e.g. photo
metadata embedded in image file itself)
 Increasing number of resource catalogs and
repositories on the web provide access to metadata
and often the resource itself
 will have seen examples for books, images, and datasets
 These repositories are indexed by search tools and
provide programmatic interfaces to allow for
resource discovery and re-use

25.  Serves same basic needs, but different scale and target of
annotation, user base, and primary use cases
II. METADATA IN DATABASE SYSTEMS
 Two main categories:
1. Structural metadata
 describes the structure
of database objects
and the relationships
between them
 commonly encoded
externally as ER-
diagrams, or internally
as summary tables
http://www.visn20.med.va.gov/VISN20/V20/DataWarehouse/Images/LabAutopsy.jpg
Example ER diagram
for VA autopsy data

26.  Serves same basic needs, but different scale and target of
annotation, user base, and primary use cases
II. METADATA IN DATABASE SYSTEMS
 Two main categories:
2. Content metadata
 describes meaning of
data at a very fine
granularity
 specifies attributes of
data elements , and
rules for recoding their
values
 encoded internally or
externally as ‘data
dictionaries’
Example of a data set that needs a dictionary to interpret

27.  The notion of a ‘data element’ obtains a more precise meaning
and specification in the context of a database.
 elements can be specified at finer granularity in a databases holding
structured data in a controlled operational system
 Conceptually, a data element is comprised of a concept and a
value domain
 concept = the subject of the data recorded for a given element
 value domain = the defined value set for how that data is recorded
 Example: PT_ETHNIC
 concept = patient ethnicity
 value domain = [ E1 (caucasian), E2 (hispanic/latino), E3 (african),
E4 (asian), E5 (mixed) ]
DATA ELEMENTS

28.  Provide detailed metadata about data elements
 element identifiers and name(s)
 definitions and descriptions
 value constraints
 data type
 default value
 length
 allowable values
 value frequency (mandatory or not)
 provenance and tracking
 version number, entry and termination dates
 indicate source table(s)
 mappings to elements in other schema dictionaries
DATA DICTIONARIES

29. DATA DICTIONARIES
http://library.ahima.org/xpedio/groups/public/documents/ahima/bok1_048618.pdf
Simple
example
of a data
dictionary

30.  Key Functions
 unambiguous and shared understanding of the data by all users
(administrators, analysts, and clients)
 consistent data representation and manipulation
(addition, extraction, aggregation, and transformation)
 maintenance of the data model
 data integration, exchange, and re-use
 Encoding
 as an external document and/or represented as a table in the
database itself
DATA DICTIONARIES

31. 1. Clear and thorough element definitions and value
set explanations are key
2. Give persistent identifiers to data elements
3. Map data elements to community standards where
possible
 common data elements (CDEs)
4. Specify value sets in terms of open controlled
vocabularies CVs where possible
5. Provide notes and guidelines for context of use
6. Make dictionary easily accessible to all users
DATA DICTIONARY
BEST PRACTICES

32.  As research moves toward 'big data‘, information from
diverse sources is being shared and aggregated for
analysis.
 A major challenge for managing this data is the diversity of
ways that a given idea can be described in data elements
 Sex/gender definitions can be based on genetics, phenotype, or self-
identification. Values can be recorded as local codes, abbreviations,
full labels, or community vocabularies.
DATABASE METADATA
INTEROPERABILITY

33.  The Common Data Element (CDE) movement aims to
address this problem by providing standardized data
elements that can be re-used across medical datasets
 CDEs are
 owned, managed, & curated by single authority (NINDS, NCI)
 stored and managed in large repositories called CDE registries
 available for diverse areas of clinical practice and research,
and at very fine granularity
 larger repositories hold up to 50,000 elements available
 CDEs serving as a foundation for interoperability across
data systems
COMMON DATA ELEMENTS (CDE )s

34.  Metadata registries that collect common data elements for a
defined domain
 Resemble large scale data dictionaries, but with key
differences:
 Exposed in searchable public repositories with additional
services to promote extraction and re-use
 Coverage is wider as they are used across different domains
and systems
 Metadata element descriptions are far richer to support
discovery, provenance, versioning, mappings, meta-modeling
 The NIH maintains a portal to information about existing CDE
initiatives, registries, and tools (http://www.nlm.nih.gov/cde/)
CDE REGISTRIES

35.  Houses >20,000 CDEs
 “Core” element set covers general concepts in medical domain
 patient demographics, medical history, assessment & examinations,
treatments & interventions, outcomes, and study protocol
 “Supplementary” sets covering specific diseases/research areas
 spinal injury, brain injury, epilepsy and stroke, Parkinson’s disease, ALS
 Metadata schema captures 30 element attributes
 this expanded set of attributes supports use cases of enabling
discovery and community re-use across different implementations
 Portal has search functionality and support for
generating clinical forms (CRFs) with CDE mappings
embedded in collected data
NINDS CDE REGISTRY
http://www.commondataelements.ninds.nih.gov/

36.  The National Cancer Institute cancer Data Standards Registry
(caDSR) is the largest and most widely used CDE registry
 >50,000 total elements
 Integrates CDEs from several initiatives under a unified model
and technical infrastructure
 Broad and deep coverage to fine granularity (as with NINDS)
 Metadata model is VERY complex
 captures >100 distinct attributes describing each data element in the
registry (vs 30 for NINDS)
 implements a complex conceptual model based on the
ISO/IEC 11179 metadata registry standard
 decomposes data elements into component parts that are
mapped to NCI thesaurus terms (formal encoding of semantics)
NCI DSR CDE REGISTRYca
https://cdebrowser.nci.nih.gov/CDEBrowser/

37. DSR CONCEPTUAL MODEL
1. To understand the data element table and explain
why it is so expansive
2. Follows a standard for database metadata
registries called ISO11179
 commonly implemented in other efforts you may
encounter
 e.g. the Clinical Data Interchange Standards Consortium
(CDISC), which has similar goals as the caDSR but across a
broader domain
3. Is the basis for semantic mappings to ontologies
such as the NCI thesaurus which are an important
feature of the model
ca

38. Data Element
Concept Value Domain
Value
Represen-
tation
Valid
Values
Class Property
DSR CONCEPTUAL MODELca

39. Data Element
PT_GENDER_CODE
Concept
‘patient gender’
Class
‘person’
Property
‘gender’
CONCEPT ELEMENT MAPPING
Concept = idea represented by the
data element, described
independently of a particular
representation
Class = a set of real world objects
with shared characteristics
Property = a characteristic common
to all members of an class

40. Data Element
PT_GENDER_CODE
Concept
‘patient gender’
Class
‘person’
C25190
Property
‘gender’
C17357
Class and property concepts
are mapped to NCI taxonomy
terms to formally encode
their semantics
Class Mapping
• person = C25190
Property Mapping
• gender = C17357
CONCEPT ELEMENT MAPPING

41. Data Element
PT_GENDER_CODE
Value Domain
VALUE DOMAIN MAPPING
Value domain = a set of attributes
describing representational
characteristics of instance data
Value Representation = type of
value the data represents
(along different dimensions)
Valid Values = the actual allowed
values for a given value domain
Value Rep.
‘person’,
‘gender’, ‘code’
Valid Values
‘0’,’1’,’2’,’9’

42. Data Element
PT_GENDER_CODE
Value Domain
Value Rep.
‘person’,
‘gender’, ‘code’
Value Representation Mappings
‘person’ = C25190
‘gender’ = C17357
‘code’ = C25162
Valid Value Mappings
0 = unknown C17998
1 = female gender C46110
2 = male gender C46109
9 = unspecified n/a
VALUE DOMAIN MAPPING
Valid Values
‘0’,’1’,’2’,’9’

43. Concept Value Domain
Value
Represen-
tation
Valid
Values
Class Property
"SEMANTICALLY UNAMBIGUOUS
INTEROPERABILITY"
 Semantic mappings of these four elements can
support more sophisticated search and analysis
 Computational tools can leverage logic in the NCI
hierarchy for query expansion and data aggregation

44.  The structure of the NCI taxonomy supports synonym
and hierarchical query expansion
LEVERAGING SEMANTICS
 User searches ‘cancer
biology’ to view all CDEs
related to this concept.
 The query is expanded
(1) to include any
children of this term in
the taxonomy, and
(2) to include elements
with text matching any
synonym of cancer in
the taxonomy
NCI Thesaurus
‘Cancer Biology’
branch

45. Strategies
Map elements in local data dictionaries to CDEs
 Parkinson’s Disease Biomarkers Program (PDBP) data dictionary
 NINDS registry form builder
Build libraries of re-usable, pre-fabricated forms with
embedded CDE metadata
 NINDS Case Report Form (CRF) library
 medical-data-models.org forms
Initialize software with CDEs so that electronic forms
automatically carry mappings when they are
generated
 caDSR registry and CDISC tools
CDE IN PRACTICEs

46.  CDEs standardize data elements for use across multiple systems
 Available in registries that vary in size and complexity
 some resemble simple data dictionaries with expanded attributes to
support discovery and provenance (NINDS)
 some are implemented with complex conceptual models and semantic
mappings (caDSR)
 Tools and standards supporting practical application exist but are
not yet state of the art
 Worlds collide: the intersection of metadata for web resources and
database systems
 CDEs represent discoverable web resources, that are used in the context of
data collection and description in database systems
 Each registry defines a metadata framework/schema for a given domain
CDE SUMMARY

47.  Promote standardized and systematic data collection
 Improve data quality and consistency
 Facilitate data sharing and integration
 Reduce the cost and time needed to develop data
collection tools
 Improve opportunities for meta-analysis comparing
results across studies
 Increase the availability of data for the planning and
design of new trials
BENEFITS OF CDEs

48.  Data elements across efforts are not well aligned
 Tooling support for discovery & application immature
 Limited use of community taxonomy and ontology
mappings
 Navigating complexity and redundancy
. . . of medical data itself
 many ways to calculate and represent simple and complex
measures such as tumor burden or medical prognosis
. . . of metadata elements/schemas
 thousands of elements with very nuanced meaning and use
 redundant representation poses challenges for data collection,
aggregation, and integrated analysis (even for simple measures)
CHALLENGES FOR DATA
INTEGRATION AND ANALYSIS

49. LINKS
Schema Examples:
DCMI: http://dublincore.org/documents/dcmi-terms/
IEEE-LOM: http://www.imsglobal.org/metadata/mdv1p3/imsmd_bestv1p3.html
HCLS Dataset description standard: http://www.w3.org/2001/sw/hcls/notes/hcls-dataset/
Data Dictionary Example:
http://library.ahima.org/xpedio/groups/public/documents/ahima/bok1_048618.pdf
CDE Sites:
NIH CDE Portal: http://www.nlm.nih.gov/cde/
NINDS CE Registry: http://www.commondataelements.ninds.nih.gov/
caDSR browser: https://cdebrowser.nci.nih.gov/CDEBrowser/
caDSR tools: http://cbiit.nci.nih.gov/ncip/biomedical-informatics-resources/interoperability-and-semantics/metadata-and-
models
CDEs in Practice:
PDBP gender data dictionary entry:
https://dictionary.pdbp.ninds.nih.gov/portal/publicData/dataElementAction!view.action?dataElementId=5585
NINDS form builder http://www.commondataelements.ninds.nih.gov/CRF.aspx?source=formBuilder
Downloadable forms (CRFs) from NINDs with embedded CDE links: http://www.commondataelements.ninds.nih.gov/CRF.aspx
medical-data-models.org forms https://medical-data-models.org/forms/1049
Suite of tools on caDSR site http://cbiit.nci.nih.gov/ncip/biomedical-informatics-resources/interoperability-and-
semantics/metadata-and-models