CONTENT-MINING IN SCIENCE
TheContentMine
Progress since “Hargreaves” legislation
Opportunities for UK, and Europe
Peter Murray-Rust, 2015-04-14
Workshop sponsored by Wellcome Trust
OUR TEAM
@jenny_molloy
Ross Mounce
@rmounce
Richard Smith-Unna
@blahah404
Stephanie Smith-Unna
@treblesteph
Jenny Molloy
Mark MacGillivray
@cottagelabs
Peter Murray-Rust
@petermurrayrust
Charles Oppenheim
@CharlesOppenh
Graham Steel
@McDawg
OUR MISSION
“make 100,000,000 facts from the STEM
literature open, accessible and reusable”
WHY?
http://www.nytimes.com/2015/04/08/opinion/yes-we-were-warned-about-
ebola.html
We were stunned recently when we stumbled across an article by European
researchers in Annals of Virology [1982]: “The results seem to indicate that
Liberia has to be included in the Ebola virus endemic zone.” In the future,
the authors asserted, “medical personnel in Liberian health centers should be
aware of the possibility that they may come across active cases and thus be
prepared to avoid nosocomial epidemics,” referring to hospital-acquired
infection.
Adage in public health: “The road to inaction is paved with research papers.”
Bernice Dahn is the chief medical officer of Liberia’s Ministry of Health,
where Vera Mussah is the director of county health services. Cameron Nutt
is the Ebola response adviser to Partners in Health.
THE RIGHT TO READ IS
THE RIGHT TO MINE
The Hargreaves report (UK) ,
legalised 2014, allowing
limitations and exceptions for
non-commercial content mining
for research.
The Hague decal
THE SCALE OF THE TASK
• ~ 27,000 peer reviewed journals*
• > 5,000 publishers
• ~ 3,000 new papers per day
• “costing” 15 Billion USD to publish
• Representing 500 Billion USD of research
*Ulrich’s database:
http://ulrichsweb.serialssolutions.com/login
OUR WORKSHOPS
• Shuttleworth Foundation
• Leicester Univ
• Electronic Theses and Dissertations
• Austrian Science Fund AT
• OKFest DE
• Eur. Bioinformatics Institute (x2)
• Open Science Rio de Janeiro BR
• Sci DataCon , Delhi IN
• Univ of Chicago US
• OpenCon 2014, Wash DC. US
• JISC , London
• LIBER
• Cochrane UK
• British Library
• Wellcome Trust
• WHO
OUR COLLABORATORS
• Shuttleworth Foundation
• Wikimedia/Wikidata
• Mozilla
• Open Knowledge
• LIBER
• British Library
• Wellcome Trust
• EBI (Eur. Bioinf. Inst.)
• JISC
• BBSRC
• Cochrane UK
• Open Access Button
• SPARC
• Creative Commons
• CORE
• EuropePubmedCentral
• Cambridge University Library
STRUCTURED INFORMATION
• chemical names and structures
• species
• metabolism
• phylogenetic trees
• …
INTERACTIVE DEMO
of content mining
http://chemicaltagger.ch.cam.ac.uk/
ContentMine at Cochrane UK, 2015-03-16
CLINICAL TRIALS
How to we find (mentions of) clinical trials?
Is a document a (clinical) trial?
What is the subject of the trial?
What is the methodology used? How many/long?
Does the design and practice conform to CONSORT?
What are the outcomes?
Can we extract specific re-usable information?
Who are involved? (researchers, sponsors, patients?)
Has a proposed trial been completed and reported?
COMMUNITY PROJECTS
• Clinical Trials (with Cochrane UK)
• Phyloinformatic Literature Unlocking Tools (PLUTo/BBSRC)
• EBI – MetaboLights
• Plant Sciences and farming (Cambridge, TGAC, OpenFarm)
• Crystallography Open Database (COD)
• OpenOil / OpenCorporates
METABOLIGHTS
• European Bioinformatics Institute
• database for metabolomics experiments and
derived information
• cross-species, cross-technique, structures,
biological roles, locations, concentrations
• http://www.ebi.ac.uk/metabolights/
CONTENTMINE WORKSHOPS AND
HACKDAYS
Open Science Brazil, 2014-08
Easily distributed software
Get started in 30 mins
Build application
in a day
Start simple: bagOfWords, Stemming, Regex, templates
What is “Content”?
http://www.plosone.org/article/fetchObject.action?uri=info:doi/10.1371/journal.pone.01113
03&representation=PDF CC-BY
SECTIONS
MAPS
TABLES
CHEMISTRY
TEXT
MATH
contentmine.org tackles these
What is “Content”?
Emily Sena (neuroscience.ed.ac.uk) spends
half a day digitising a diagram like this
ContentMine will soon be able to do it in 1 second
Note Jaggy and
broken pixels
NEW Bacteria must have a phylogenetic tree
Length
_________Weight
Binomial Name Culture/Strain GENBANK ID
Evolution
Rate
• CRAWL the web for scientific documents
(articles, grey literature, repositories)
• quickSCRAPE pages (text, graphics, images, data)
• NORMA-lize page to semantic form
…Open semantic science …
• MINE pages with your methods and tools (AMI)
• CAT-alogue results in searchable index
• Automate daily process (CANARY)
contentmine.org Infrastructure
quickscrape
Crawl
Feed
Norma
Index &
Transform
PDF
XML
URL
DOI
Scientific
literature
Repositories DOC
CSV
sHTML
Plugins
Regex
SequencesSpecies
Bespoke
Scrapers
XPathPer-Journal
Taggers
Per- Journal
MetadataChemistry
Phylogenetics Farming
AMI
BadHT
ML
OCR
Diagrams
Open NORMA-lized Scientific
Literature + Facts
CANARY pipeline
CAT-alogue index
POSSIBLE USES
• Indexing/searching the literature; G***** for science
• Current awareness; alerts and practices
• Extraction and re-use of facts; re-computation
• Multidisciplinary integration; co-occurrence
• Compliance with funder/institution policies
• Managing your Research Data!
• Finding similar and complementary colleagues
• Reproducibility, checking data and avoiding fraud
How to leverage Content
Mining for benefit of UK/EU
• Create UK showcase of successes in mining
• Graduate training by 3rd year UK graduate students.
• Develop EuropePMC as world resource for bio-mining
• Training/support for UK/EU libraries about Hargreaves.
• Central collection of born-digital UK theses
• Collect pre-copyright author manuscripts
• Integrate CM into Research Data Management tools
• Promote mining in all aspects of healthcare information
• Open collection of extracted scientific facts for the world

Content Mining at Wellcome Trust

  • 1.
    CONTENT-MINING IN SCIENCE TheContentMine Progresssince “Hargreaves” legislation Opportunities for UK, and Europe Peter Murray-Rust, 2015-04-14 Workshop sponsored by Wellcome Trust
  • 2.
    OUR TEAM @jenny_molloy Ross Mounce @rmounce RichardSmith-Unna @blahah404 Stephanie Smith-Unna @treblesteph Jenny Molloy Mark MacGillivray @cottagelabs Peter Murray-Rust @petermurrayrust Charles Oppenheim @CharlesOppenh Graham Steel @McDawg
  • 3.
    OUR MISSION “make 100,000,000facts from the STEM literature open, accessible and reusable”
  • 4.
    WHY? http://www.nytimes.com/2015/04/08/opinion/yes-we-were-warned-about- ebola.html We were stunnedrecently when we stumbled across an article by European researchers in Annals of Virology [1982]: “The results seem to indicate that Liberia has to be included in the Ebola virus endemic zone.” In the future, the authors asserted, “medical personnel in Liberian health centers should be aware of the possibility that they may come across active cases and thus be prepared to avoid nosocomial epidemics,” referring to hospital-acquired infection. Adage in public health: “The road to inaction is paved with research papers.” Bernice Dahn is the chief medical officer of Liberia’s Ministry of Health, where Vera Mussah is the director of county health services. Cameron Nutt is the Ebola response adviser to Partners in Health.
  • 5.
    THE RIGHT TOREAD IS THE RIGHT TO MINE The Hargreaves report (UK) , legalised 2014, allowing limitations and exceptions for non-commercial content mining for research. The Hague decal
  • 6.
    THE SCALE OFTHE TASK • ~ 27,000 peer reviewed journals* • > 5,000 publishers • ~ 3,000 new papers per day • “costing” 15 Billion USD to publish • Representing 500 Billion USD of research *Ulrich’s database: http://ulrichsweb.serialssolutions.com/login
  • 7.
    OUR WORKSHOPS • ShuttleworthFoundation • Leicester Univ • Electronic Theses and Dissertations • Austrian Science Fund AT • OKFest DE • Eur. Bioinformatics Institute (x2) • Open Science Rio de Janeiro BR • Sci DataCon , Delhi IN • Univ of Chicago US • OpenCon 2014, Wash DC. US • JISC , London • LIBER • Cochrane UK • British Library • Wellcome Trust • WHO OUR COLLABORATORS • Shuttleworth Foundation • Wikimedia/Wikidata • Mozilla • Open Knowledge • LIBER • British Library • Wellcome Trust • EBI (Eur. Bioinf. Inst.) • JISC • BBSRC • Cochrane UK • Open Access Button • SPARC • Creative Commons • CORE • EuropePubmedCentral • Cambridge University Library
  • 8.
    STRUCTURED INFORMATION • chemicalnames and structures • species • metabolism • phylogenetic trees • …
  • 9.
    INTERACTIVE DEMO of contentmining http://chemicaltagger.ch.cam.ac.uk/
  • 10.
  • 11.
    CLINICAL TRIALS How towe find (mentions of) clinical trials? Is a document a (clinical) trial? What is the subject of the trial? What is the methodology used? How many/long? Does the design and practice conform to CONSORT? What are the outcomes? Can we extract specific re-usable information? Who are involved? (researchers, sponsors, patients?) Has a proposed trial been completed and reported?
  • 12.
    COMMUNITY PROJECTS • ClinicalTrials (with Cochrane UK) • Phyloinformatic Literature Unlocking Tools (PLUTo/BBSRC) • EBI – MetaboLights • Plant Sciences and farming (Cambridge, TGAC, OpenFarm) • Crystallography Open Database (COD) • OpenOil / OpenCorporates
  • 13.
    METABOLIGHTS • European BioinformaticsInstitute • database for metabolomics experiments and derived information • cross-species, cross-technique, structures, biological roles, locations, concentrations • http://www.ebi.ac.uk/metabolights/
  • 14.
    CONTENTMINE WORKSHOPS AND HACKDAYS OpenScience Brazil, 2014-08 Easily distributed software Get started in 30 mins Build application in a day Start simple: bagOfWords, Stemming, Regex, templates
  • 15.
  • 16.
    What is “Content”? EmilySena (neuroscience.ed.ac.uk) spends half a day digitising a diagram like this ContentMine will soon be able to do it in 1 second
  • 17.
    Note Jaggy and brokenpixels NEW Bacteria must have a phylogenetic tree Length _________Weight Binomial Name Culture/Strain GENBANK ID Evolution Rate
  • 18.
    • CRAWL theweb for scientific documents (articles, grey literature, repositories) • quickSCRAPE pages (text, graphics, images, data) • NORMA-lize page to semantic form …Open semantic science … • MINE pages with your methods and tools (AMI) • CAT-alogue results in searchable index • Automate daily process (CANARY) contentmine.org Infrastructure
  • 19.
    quickscrape Crawl Feed Norma Index & Transform PDF XML URL DOI Scientific literature Repositories DOC CSV sHTML Plugins Regex SequencesSpecies Bespoke Scrapers XPathPer-Journal Taggers Per-Journal MetadataChemistry Phylogenetics Farming AMI BadHT ML OCR Diagrams Open NORMA-lized Scientific Literature + Facts CANARY pipeline CAT-alogue index
  • 20.
    POSSIBLE USES • Indexing/searchingthe literature; G***** for science • Current awareness; alerts and practices • Extraction and re-use of facts; re-computation • Multidisciplinary integration; co-occurrence • Compliance with funder/institution policies • Managing your Research Data! • Finding similar and complementary colleagues • Reproducibility, checking data and avoiding fraud
  • 21.
    How to leverageContent Mining for benefit of UK/EU • Create UK showcase of successes in mining • Graduate training by 3rd year UK graduate students. • Develop EuropePMC as world resource for bio-mining • Training/support for UK/EU libraries about Hargreaves. • Central collection of born-digital UK theses • Collect pre-copyright author manuscripts • Integrate CM into Research Data Management tools • Promote mining in all aspects of healthcare information • Open collection of extracted scientific facts for the world

Editor's Notes

  • #2 This presentation will be a quick introduction to the ContentMine software for literature scraping, normalising, and fact extraction.
  • #3 Because information is structured (some examples listed), we can aggregate similar objects and mine using a modular systematic approach.
  • #9 Because information is structured (some examples listed), we can aggregate similar objects and mine using a modular systematic approach.
  • #12 Because information is structured (some examples listed), we can aggregate similar objects and mine using a modular systematic approach.
  • #13 Can describe each collaboration, but keep this slide brief if the presentation is short.
  • #14 Can describe each collaboration, but keep this slide brief if the presentation is short.
  • #15 Can describe each collaboration, but keep this slide brief if the presentation is short.
  • #21 Can describe each collaboration, but keep this slide brief if the presentation is short.
  • #22 Can describe each collaboration, but keep this slide brief if the presentation is short.