Textmining activities at BioHackathon 2010

1,448 views
1,378 views

Published on

Summary of the activities developed by the text mining task force during the BioHackathon 2010

http://hackathon3.dbcls.jp/wiki/TextMining

Published in: Education, Technology
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,448
On SlideShare
0
From Embeds
0
Number of Embeds
5
Actions
Shares
0
Downloads
5
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

Textmining activities at BioHackathon 2010

  1. 1. Semantic Textmining Goals and achievements BioHackathon 2010
  2. 2. Team members • Hammad • Matthias • Venkata • Heiko • YAMAMOTO-san • Alberto
  3. 3. Original Proposal • Integration of text mining results – Reflect / Whatizit / Medie – Results as triplets • URI and predicates – Implementation with SADI – Result presentation using aTag • Explore relations • Interfaces
  4. 4. The work done • Integration of text mining results – Reflect / Whatizit / Medie – Results as triplets • URI and predicates – Future BioPython module and REST service • Explore relations – Sesame endpoint – Biogateway – ARQ for federated queries • Interfaces – Result presentation using aTag – Exhibit faceted interface
  5. 5. http://whatizit.neurocommons.org/
  6. 6. http://whatizit.neurocommons.org/
  7. 7. RDF schema for TM <rdf:Description> <rdf:type rdf:resource="http://rdfs.org/sioc/ns#Item"/> <sioc:about rdf:resource="http://www.ncbi.nlm.nih.gov/pubmed/9002550"/> <sioc:content>SBMA</sioc:content> <sioc:topic rdf:resource="http://purl.uniprot.org/uniprot/P10275"/> <rdfs:seeAlso rdf:resource="http://www.ncbi.nlm.nih.gov/pubmed/9002550"/> </rdf:Description>
  8. 8. http://reflect.ws
  9. 9. MEDIE and Enju APIs • MEDIE is an intelligent search engine to retrieve biomedical correlations from MEDLINE, based on indexing by Natural Language Processing and Text Mining techniques. • Enju is a syntactic parser for English.
  10. 10. Medie XML output http://www-tsujii.is.s.u-tokyo.ac.jp/medie/dbcls.cgi?pmid=19116711
  11. 11. Enju XML output http://docman.dbcls.jp/medieconv?pmid=17551671
  12. 12. http://togows.dbcls.jp/entry/pubmed/pmid.ttl http://www.uniprot.org/uniprot/P12345.rdf
  13. 13. Workflow Whatizit Reflect Medie Pubmed XML XML XML RDF RDF TogoWS RDF Uniprot
  14. 14. Substance A Interacts with Receptor B Region C axonal projections brain region D Region D aversive stimuli Interlink these entities with taxonomies & ontologies
  15. 15. TMOntology
  16. 16. http://hackathon3.dbcls.jp/wiki/TextMining

×