Text Mining for Scholarly Communications and Repositories Joint Workshop

Venue: Manchester Interdisciplinary Biocentre, University of Manchester
Dates: 28-29th October 2009
Audience: researchers, information management professionals, librarians, text miners, repository providers, publishers, policy makers, JISC service representatives.
Aim: To examine the issues, challenges and priorities associated with integrating text mining technologies in applications to support scholarly communication and repository initiatives.
Organisers: Sophia Ananiadou (NaCTeM), Liz Lyon (UKOLN)

Programme of talks
External posters and demos
NaCTeM demos

Programme of Talks

Day 1 (Wednesday 28th October)
14.40Introduction and Overview (Dr. Lyon)

Scholarly communications and Text Mining : a view of the landscape (Prof. Ananiadou)
Professor Sophia Ananiadou, Director, NaCTeM,
Dr Liz Lyon, Director, UKOLN
15:00Using text-mining for discovery and data integration: literature resources at the EBI and the UKPMC projectDr Johanna McEntyre, European Bioinformatics Institute, Cambridge
15.30Mining and meaning in the chemical sciencesRichard Kidd, Royal Society of Chemistry
16.30Approaches to automated metadata extraction : FixRep ProjectEmma Tonkin, UKOLN, University of Bath
17:00Institutional Repository SearchVic Lyte, MIMAS
17.30Text Mining to support systematic reviews in the social sciencesJames Thomas, Assistant Director, EPPI-Centre, Institute of Education, University of London

Day 2 (Thursday 29th October)
09.30Keynote: eScience and Semantic ComputingProfessor Tony Hey, Corporate Vice President, Microsoft External Research
11.00Open UpRafael Sidi, Vice-President Product Management, Elsevier
11.30A Service Perspective: Unlocking metadata to enhance discoverability and contextPeter Burnhill, EDINA
12.00Citations and SentimentSimone Teufel, Computer Laboratory, University of Cambridge

External Posters and Demos

PresenterPoster TitleDemo Title
Dr. Maria Liakata, Aberystwyth UniversitySAPIENT Automation Project: Automatically Recognising Core Scientific Concepts in full text papersSAPIENT system
Kate Byrne, Edinburgh UniversityPopulating the Semantic Web with Relations from TextGeo-parser - finding spatial references in text and putting them on the map
Dr Steve Pettifer, University of ManchesterUtopia networks

NaCTeM Demos

KLEIOKLEIO offers a more convenient and intuitive tool for browsing document repositories by using textual and metadata search. It integrates various text mining components to offer textual and metadata search and enhanced functionality in terms of acronym expansion and interactive ranking.
FACTA+FACTA is a tool that helps discover associations between biomedical concepts contained in MEDLINE articles. It allows the user to provide a flexible query (e.g. keywords or Boolean combinations of concepts) and retrieves the documents that match and the associations between the query term and concepts in a highly interactive manner.
MEDIEThe MEDIE is an intelligent search engine for finding biological events from MEDLINE and discovering interactions between biomedical entities. The service uses advanced natural language processing technologies and a novel search engine for the accurate retrieval of concepts. It identifies sentence level biological relationships and map relevant facts to queries presented by the user
TerMineTerMine is a Term Management System which identifies key phrases in text. Existing terminological resources and scientific databases cannot keep up-to-date with the growth of neologisms. In response to this, TerMine offers domain-independent method for term recognition in documuments.
AcroMineAcromine is an abbreviation dictionary automatically constructed from the whole of MEDLINE. Abbreviations result from a highly productive type of term variation which substitutes fully expanded terms (e.g., retinoic acid receptor alpha) with shortened term-forms (e.g., RARA). Even though no generic rules or exact patterns have been established for dealing with abbreviation creation, abbreviations often appears in documents without the expanded form explicitly stated. Thus, an abbreviation dictionary is necessary for advanced text-mining tasks to establish associations between abbreviations and their expanded forms.