Prof Sophia Ananiadou - research
I have published over 165 peer reviewed publications in journals and conferences. My main contributions are in the area of natural language processing, and in particular computational terminology and biomedical text mining. My work in computational terminology and term recognition led to the development of the C-value method for automatic term recognition which has been adopted as a standard method internationally. It is now used by one of the most widely accessed text mining services of the National Centre for Text Mining. In 2006, I edited the first book on biomedical text mining, which is also used as a textbook.
I have published in high impact factor journals such Trends in Bioinformatics (IF 6.909), Nucleic Acids Research (IF 7.479), Bioinformatics (IF 5.039), Briefings in Bioinformatics (IF 4.415); BMC Bioinformatics (IF 3.49); PLoS One 4.351. I have edited 2 special issues of BMC Bioinformatics and one of the Journal of Biomedical Informatics (IF 2.000). Citation index (WoK) had a six fold increase since 2006 and a two fold increase since 2009. The average citations per item increased from 3.15 in 2009 to 7.50 in 2011.
In Natural Language Processing (NLP), high quality conferences are considered of equal value to journal publications, especially given the limited number of specialized journals in the field. The prime conferences in NLP have typically an acceptance rate of 15% to 20%; these include Association for Computational Linguistics (ACL), North American Chapter of the Association for Computational Linguistics - Human Language Technologies (NACCL-HLT), European Chapter of the Association for Computational Linguistics (EACL), International Conference on Computational Linguistics (COLING) and Empirical Methods of Natural Language Processing (EMNLP). I have over 25 publications in these conferences. I am an associated editor of the journal Computational Intelligence, impact factor 5.378
My H-Index stands at 23 (Google Scholar).
In 2010, I supervised the team NaCTeM in the critical assessment of information extraction in biology BioCreAtIvE III protein-protein interaction (PPI) challenge and achieved the best performance, in what was considered the most challenging task, the Interaction Method Task (IMT). This involves automatically detecting experimental techniques used in research articles that support given PPIs. Such detection is crucial not only for the correct annotation of experimentally determined protein interactions but also for other annotations such as evidence codes in Gene Ontology and assigning other controlled vocabulary terms to an article. Among systems submitted by eight international teams, NaCTeM's yielded the best overall performance as measured by a range of evaluation metrics.