The Missing Link-The Evolving Current State of Linked Data for Serials-Lauruhn
description
Transcript of The Missing Link-The Evolving Current State of Linked Data for Serials-Lauruhn
Linked Data at Elsevier
Michael P Lauruhn
The Missing Link: The Evolving Current State of Linked Data for Serials
NASIG 28th Annual Conference
June 7, 2013
@MikeLauruhn
@ElsevierLabs
2
• Create greater online engagement with our content and platform
• Create additional usage in journals and books, through interactive use and downloads
• Semantically enrich content and increase value of discovery services compared to the similar content at other platforms
• As a publisher, add value and improve our status as a partner in research
The challenge for publishers
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
3
Our Approach
• Expose asset and subject metadata as linked data in Web pages to help discovery
• Use linked data principles while using our current content production workflow
• Use of standard vocabularies, taxonomies, ontologies and entity lists whenever possible
• Leverage partners for content enhancement and knowledge organization
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
Smart Content: Semantic Enhancements for Scientific Publishing
Concepts:
Metadata,
Entities,
Relationships
Applied Smart Content
• Faceted search & browse
• Ontology-driven navigation
• Task-specific results
• Personalized/localized results
• Question answering
• Topic pages
• Social network maps
• Geolocation maps
• Data mashups
• Text mining reports
Multimedia
Text
Data
Elsevier
content
Related
Elsevier
content
and data
Linked data from
partners and the
Web
• Tag clouds
• Heatmaps
• Streamgraphs
• Scatterplots
• Time series
• Animations
Better discovery
Better understanding
Actionable, persuasive knowledge
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
5
What metadata can we collect & link
Asset Metadata
Bibliographic Metadata
Entities
Relations
Citations
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
6
Entities Example: EMMeT
Core Semantic Types:
Diseases
Anatomy
Clinical Findings (includes symptoms)
Procedures
Drugs
Sources:
EMMeT
SNOMED CT
FMA (Foundational Model of Anatomy)
GO (Gene Ontology)
MeSH
NCIt (National Cancer Institute thesaurus)
Elsevier Merged Medical Taxonomy
Entities
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
7
<skosxl:literalForm xml:lang="en-US">Diabetes Mellitus</skosxl:literalForm> <ebs:usageFlag rdf:resource="http://data.elsevier.com/EMMeT/Flags/MedicalName"/> ... <skosxl:literalForm xml:lang="en-US">Diabetes</skosxl:literalForm> <ebs:usageFlag rdf:resource="http://data.elsevier.com/EMMeT/Flags/ConsumerFriendlyName"/> ... <skos:notation rdf:datatype="http://data.elsevier.com/vocabulary/EMMeT">177824</skos:notation> <skos:notation rdf:datatype="http://dbpedia.org/resource/UMLS">C0011849</skos:notation> ... <rdf:type rdf:resource="http://data.elsevier.com/EMMeT/SemTypes/DiseaseOrSyndrome" /> <skos:broader rdf:ID="Relation-34359" rdf:resource="http://data.elsevier.com/vocabulary/EMMeT/Concept/48543"/> ... <skos:narrower rdf:ID="Relation-9812" rdf:resource="http://data.elsevier.com/vocabulary/EMMeT/34565"/> ... <emsem:hasSymptom rdf:ID="Relation-99999"
rdf:resource="http://data.elsevier.com/vocabulary/EMMeT/Concept/53425"/>
… Abnormal Sense of Taste
Abnormal metabolic state in diabetes mellitus
Disorders of endocrine system
EMMeT – Elsevier Merged Medical Taxonomy
Term example: Diabetes Mellitus
Relations
Entities
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
8
Rivastigmine, a cholinesterase inhibitor, has been used to
treat delirium in elderly patients with stroke. 1 A
biologically plausible premise—that impaired cholinergic
transmission might either cause or worsen delirium—led
to a randomised, placebo-controlled, double-blind trial by
Maarten van Eijk and colleagues 2 in The Lancet in which
they added rivastigmine or placebo to usual treatment of
patients in intensive care. The trial was halted at 104
patients by the drug safety and monitoring board (DSMB)
because of increased mortality (12/54 in the rivastigmine
group, 4/50 in the placebo group; p=0·07) and a worse
outcome. The rivastigmine group …
Linked Data Repository (LDR): Warehouse for Smart Content Enhancements
Delirium treatment: An unmet challenge Title
Drug Clinical finding
• Enhances extracted knowledge of Elsevier
assets by interlinking data with related sources
of medical and scientific content and data.
• Optimized for high-volume read-write for use
by end-user products.
• Provide service layer APIs for ease of
integration.
Disease
ATC: N06DA03 Drug: Rivastigmine
med:diseases Delirium med:drugs Rivastigmine
Elsevier
Trial: NCT00623103 Intervention: Rivastigmine Condition: Delirium
LinkedCT Trial: NCT00623103 Serious Adverse events: Atrial fibrillation
owl:same as
owl: same as
foaf:page
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
9
Some Examples
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
10
Linked Data Prototype for The Lancet
Special featureThe Lancet special issue: ―Stillbirths‖
(Vol 377; Number 9774; April 14, 2011)
Creation LDR-enabled interactive application
using:
• The Lancet content
• Datasets from The Lancet editorial staff
• Datasets from World Bank
• EMMeT subject tagging
• Geo locations & Map
The Lancet and World Bank datasets
loaded into the LDR as triples.
EMMeT tagging results of articles
loaded into LDR.
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
11
Trend Analysis Of Special Health Topics: Stillbirths
http://ldr.elsevier.com/lancet
Countries color coded
by Lancet data on
stillbirth rate per 1000
births.
Bubbles represent
selected data from
World Bank
(GDP per Capita)
Narrow search by
identifying articles with
most relevant concepts
related to stillbirth.
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
12
http://ldr.elsevier.com/ukgov/map.html
Regions color coded
based on avg. cost of
Tamoxifen treatment
per 1000 people.
Bubbles represent
the cost of treating
side effects of Tamoxifen.
Example; North
Yorkshire
Additional cost for
Toremifene = £357k
If we assume an
adverse effect costs
£300 per patient.
Tamoxifen = £1.48m
Toremifene =
£103k
Savings Potential
>£1m by prescribing
a seemingly more
expensive drug
Healthcare Analytics: Breast Cancer Treatment in the UK
Data from
data.gov.uk,
Elsevier articles,
& Elsevier
Pharmapendium
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
13
Neuroscience Research: Improved Access To Data and Better Tools
CHALLENGE Neuroscience is a highly interdisciplinary field requiring analysis on vast amounts of content from many different sources:
• Foundational information is hard to find • Need to identify articles with specific methods or related experiments • Analyze anatomy, connectivity, and gene expression data from different sources • Sources use different ontologies
`
Unpaired midbrain region situated in the ventromedial portion of the reticular formation. The VTA is medial to the substantia nigra and ventral to the red nucleus, and extends caudally ….. Anatomy Connectivity Expression
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
14
• Increasing acquisition of data and text analytics capabilities
• Shifting dependence from partners to in-house resources for metadata creation and data modeling
• Innovation in new knowledge organization systems – taxonomy for discovery
– ontology for understanding and integration
• Emergence of shared infrastructure based on linked data principles
Trends within Elsevier today
Copyright © 2013 Elsevier, Inc. | All Rights Reserved
Thank you
Michael Lauruhn
@MikeLauruhn
@ElsevierLabs
Copyright © 2013 Elsevier, Inc. | All Rights Reserved