The challenge of collaboration across communities

I wonder how many of us relate to Gary Larson's Ginger

Original questions

What kind of controlled vocabulary is used in your application? Could it be shared with the community?

• No; None

What data models have been developed in your project? Could they be shared with the community?

• RDF; Yes!

Is there an automated data validation process in your application? Could it be shared with the community?

• No; No

What kinds of analysis tools are used in your project? Could they be shared with the community?

• None; No

What are the challenges in integrating clinical data with basic research data and how have you resolved them?

• RDF facilitates the integration of structured and semi-structured data.

What standards are you using in your projects (CRFs, file formats, controlled vocabularies, file exchange, etc.)?

• We develop standards - URI, XML, RDF, OWL that provide the foundation for representing, integrating data on the Web

Semantic Web - Many things to many people

The World Wide Web Consortium

• International consortium directed by Tim Berners-Lee
• Mission: "Lead the Web to its full potential"
• Hosts: MIT, ERCIM, Keio University
• Defines Web standards: HTML, XML, Web Services, Semantic Web, Mobile Web, WAI, I18N, etc.
• W3C track record: building infrastructure to address technical and social needs of the Web

Semantic Web

• Data Integration across application, organizational, community boundaries
• Reduces the technical and social costs for effective integration of networked data at various scales

How does it work?

• Apply power of URIs to concepts of relational data
• Model real things, not documents or database tables

Semantic Web: Identifying Concepts

• Don't say "colour" say <http://example.com/2002/std6#col>

The relational database

The element of the Semantic Web

• Can be encoded in XML
• Simplicity and mathematical consistency
• This is called Resource Description Framework (RDF)

Semantic web includes tables,...

...trees

 

 

... everything

Building blocks

... enabling many things

Wrapping, Enhacing the Existing Web

• Web Evolution not Revolution
• Exposing data hiding in documents, servers and databases
• Machine Processible data on the Web

Applications connected by concepts

Enabling Data Integration

Facilitate the management, combination of data and services Piggy-Bank: Personal information management Semantic-Bank: Collaborative information management Solvent: Interactive Tool for helping exract data from particular web sites

Fractal Web of concepts

• Across boundaries of scale -- personal, group, corporate, community, global
• Varying access levels
• Tension between local and global standards
• Society is a fractal tangle, so must the Semantic Web.

Semantic Web Technologies

• Semantic Web foundation specifications RDF, RDF Schema and OWL are W3C Recommendations as of Feb 2004 (press release and testimonials)
• Standardization work is underway in Query, Best Practices and Rules
• Community work is underway in Semantic Web Health Care and Life Sciences
• Gloal of moving from a Web of Document to a Web of Data

Semantic Web Standards

• Open standards, commercial and open source and tools - technologies for modeling real world things; sharing these models across the Web.
• Focus is on Web Evolution
• Web Evolution causing a quiet data revolution

Random sampling

• Nokia
• Oracle
• Vodafone
• IBM
• Adobe
• HP
• etc.

Equally impressive Open Source stories

Common Themes

• No "one schema" that can be used for describing everything
• No "right way" for describing / organizing anything
• Desire for Recombinant Data
• Importance of "Partial Understanding"
• Things change, plan up front for it
• Need to free data from the application that created it
• The value in "as needed" data integration
• Big wins come from many little ones
• The power of links - network effect
• Open-world, open solutions are cost effective

Social Demands on Semantic Web Architecture

Requirements

• Security
• High-granularity access control
• Privacy
• Intellectual Property management

W3C has faced these on the Web today:

• Platform for Privacy Preferences
• XML Signature, Encryption, Key Management
• DRM, in progress

Semantic Web / Life Sciences

W3C Semantic Web for Life Sciences Workshop, Oct 2004

• Too much information - no one can know everything
• No one standard can represent everything
• Different scales
• Different perspectives
• The problem of naming
• Information sharing policies (personal, group, corporate, public)
• Things change
• Big wins come from many little ones

W3C Health Care and Life Sciences Group

Chairs: Eric Neumann, Teranode and Tonya Hongsermeier, Partners Healthcare

• Share use cases, applications, demonstrations, experiences
• Applying tools, developing guidelines for exposing collections in RDF
  • GRDDL; A mechanism for connecting XHTML (e.g. microformats) and XML dialects with the Semantic Web.
  • Simple Knowledge Organization System (SKOS)
  • SPARQL ; Common protocol and interface to expose databases / application data as RDF - "Join the Web". Data Access working group
• Building / extending (where appropriate) core vocabularies for data integration

Lots of demos and applications

For those interested in learning more

• Partners Healthcare Systems on 'Clinical Knowledge Management'
• Agfa on 'Connected Knowledge' on Clinical Trials
• Active Semantic Electronic Medical Record, an application of Active Semantic Documents in health care - Semagix
• BioDash using Haystack; end user steering of decentralized data - IBM, Teranode
• Language & Computing's Knowledge Discovery Platform accelerates drug development through knowledge extraction and semantic integration - Language & Computing

Meeting Goals - Revisited

Identify areas of potential collaboration, synergy and reusable resources within the different DAIT funded bioinformatics and data management programs

• Too early to say :) Ask me tomorrow!

Meeting Goals - Revisited

Identify ways to bridge the gaps between clinical and mechanistic data exchange

• Expose services via the Web; use RDF / OWL for data representation

Meeting Goals - Revisited

Facilitate DAIT's clinical and basic science research agendas

• Model the real world; Make core scientific resources that faciliate basic science availiable via the Semantic Web. Develope core vocabularies / ontologies that help bind together basic science and research.
• Think about machine readble persistence policies, licenses for data (e.g. ScienceCommons)
• Invest in the Web - "think globally, act locally"
• Choose Open Standards; your data is too important not to
• Shaping requirements for RDF Forms? (GRDDL'ing XForms)

Meeting Goals - Revisited

Identify workgroups for further standards development among the different DAIT programs

• W3C (in general)
• W3C Health Care and Life Sciences Group (specifically)

More Information

• Slides are available at http://www.w3.org/2006/Talks/0216-semweb-em/
• W3C - http://www.w3.org/
• Semantic Web Home Page - http://www.w3.org/2001/sw/
• Me - Eric Miller, em@w3.org