spacer

Rebholz Group

The group focuses on extraction of facts from scientific literature in molecular biology. This is mainly based but not limited to matching of language patterns. In addition we do research on the disambiguation of semantic types, e.g. proteins, genes, species, drugs, and on automatic methods to identify language patterns. Both approaches require machine learning expertise and domain knowledge. The group has strong experience in Natural Language Processing methods (NLP) in the biomedical domain, and has applied its methods to different problems, e.g. identification of protein-protein interactions, extraction of acronyms and identification of mutations. All solutions are implemented as stand-alone modules, which can be combined and used to process text data in a pipeline. Existing modules are made available to the public through EBI's central services ('Whatizit', now as SOAP Web Service).

In 2005 the team has developed a special information retrieval engine ('EBIMed'), which processes Medline abstracts and delivers the analysis to the user for fast exploration. During the analysis terms from a large set of terminologies, e.g. UniprotKB/Swiss-Prot protein names, gene ontology terms, drugs and species, are identified. In addition acronyms referring to protein names are disambiguated. This solution shows, how information retrieval and information extraction can be combined, how pattern matching and disambiguation modules can be combined in a pipeline of modules, and how research work contributes to a powerful solution for daily use.

Currently the group is involved in a number of initiatives that prepare text mining solutions for their use in the public (refer to Whatizit). One initiative started at the ISMB 2006 in Fortaleza, Brasil, and focuses on the development of solutions that eases the flow of information provided through the scientific literature into other electronic data repositories.

Rebholz group is headed by Dietrich Rebholz-Schuhmann.

For Text Mining Support please go to: http://www.ebi.ac.uk/Rebholz-srv/
For enquiries, problems with our software and membership to mailing lists, please email to: textmining-support@ebi.ac.uk.

Date: November 8/10, 2009
Venue: Jeju Island, South Korea
Registration Deadline: ---

LBM is a unique interdisciplinary forum that brings together researchers in biology, chemistry, medicine, public health and informatics to discuss and exploit cutting edge language technology. Language, in its many forms, is the universal means to represent, question, and convey knowledge. The automation of language processing advances our ability to discover new knowledge... more

Date: 5th - 6th October , 2009
Venue: EBI, Hinxton, Cambridge, UK
Registration Deadline: Till September 10th, 2009

The European Bioinformatics Institute (EBI) and the National Centre for Text Mining (University of Manchester) are organising a joint training event at the EBI. The purpose of this event is to teach basic techniques in information retrieval (IR) and information extraction (IE) in the biomedical domain and to give hands-on training on existing solutions provided by the two centres...more
spacer
spacer