Table 2.
Participating teams
| Institution/department | NER tool summary | Primary contact <email address> | Web service URL(s) |
|---|---|---|---|
| National ICT Australia/Victoria Research Laboratory (Melbourne, Australia) | Gene, chemical, disease and action term: dictionary-based lookup using ConceptMapper (22), with custom dictionary construction algorithms and parameter tuning. | Andrew MacKinlay <Andrew.MacKinlay@nicta.com.au> | To be determined |
| Wuhan University (Wuhan, Hubei, China) | Gene, chemical and disease: a proprietary dictionary-based NER tagger | Cheng Sun <whucsnlp@gmail.com> | http://nlp.whu.edu.cn/Bio_NER/gene |
| http://nlp.whu.edu.cn/Bio_NER/chem | |||
| http://nlp.whu.edu.cn/Bio_NER/disease | |||
| http://nlp.whu.edu.cn/Bio_NER/action_term | |||
| Action term: LIBSVN (23) | |||
| University of Applied Sciences of Western Switzerland, Geneva/BiTeM Group, Information Science Department (Geneva, Switzerland) | Gene: NormaGene (24) | Dina Vishnyakova <dina.vishnyakova@unige.ch> | http://pingu.unige.ch:8080/Toxicat/rest/gene/post/ |
| Diseases and chemicals: ad hoc keyword recognizer based on the controlled vocabulary dictionaries provided by CTD | http://pingu.unige.ch:8080/Toxicat/rest/chem/post/ | ||
| http://pingu.unige.ch:8080/Toxicat/rest/disease/post/ | |||
| http://pingu.unige.ch:8080/Toxicat/rest/action_term/post/ | |||
| Action Terms: GOCat (25), a proprietary thesaurus and machine learning-based system | |||
| University and University Hospitals of Geneva/Division of Medical Information Sciences (Geneva, Switzerland) | |||
| SIB Swiss Institute of Bioinformatics/SIBtex (Geneva, Switzerland) | |||
| University of Zurich/Institute of Computational Linguistics (Zurich, Switzerland) | Gene, chemical and disease: Ontogene (26), a lexical lookup and machine learning-based entity tagger. | Fabio Rinaldi <fabio.rinaldi@uzh.ch> | http://kitt.cl.uzh.ch:8081/gene |
| http://kitt.cl.uzh.ch:8081/chem | |||
| http://kitt.cl.uzh.ch:8081/disease | |||
| http://kitt.cl.uzh.ch:8081/action_term | |||
| Action terms: a document classifier-based entity tagger | |||
| Academia Sinica/Institute of Information Science (Taipei, Taiwan) | Gene: BioC-GN, a proprietary machine learning- and dictionary-based NER module. | Hong-Jie Dai <hjdai@tmu.edu.tw> | http://bws.iis.sinica.edu.tw/BioC_GN/RestServiceImpl.svc/gene |
| Yuan Ze University/Department of Computer Science & Engineering (Taoyuan, Taiwan) | |||
| Taipei Medical University/Graduate Institute of BioMedical Informatics (Taipei, Taiwan) | |||
| National Tsing-Hua University/Department of Computer Science (HsinChu, Taiwan) | |||
| National Central University/Department of Computer Science and Information Engineering (Zhongli City, Taiwan) | |||
| National Cheng Kung University/Department of Computer Science and Information Engineering (Tainan, Taiwan) | Gene, chemical, disease and action term: adapted version of CoINNER, a proprietary dictionary/conditional random fields-based NER tagger | Hung-Yu Kao <hykao@mail.ncku.edu.tw> | http://140.116.245.192:8080/coinner/gene |
| http://140.116.245.192:8080/coinner/chemical | |||
| http://140.116.245.192:8080/coinner/disease | |||
| http://140.116.245.192:8080/coinner/action_term | |||
| National Cheng Kung University/Department of Computer Science and Information Engineering (Tainan, Taiwan) | Gene, chemical, disease and action term: adapted version of GCDA, a proprietary dictionary- and search engine-based NER tool that integrates OSCAR4 (27), LingPipe (28), MetaMap (16) and Lucene (29) | Jiun-Huang Ju <jujh@iir.csie.ncku.edu.tw> | http://rose.csie.ncku.edu.tw:8080/rest/gene |
| Jung-Hsien Chiang <jchiang@mail.ncku.edu.tw> | http://rose.csie.ncku.edu.tw:8080/rest/chem | ||
| http://rose.csie.ncku.edu.tw:8080/rest/disease | |||
| http://rose.csie.ncku.edu.tw:8080/rest/action_term | |||
| Mayo Clinic/Department of Health Sciences Research (Rochester, MN, USA) | Gene, chemical and disease: a proprietary dictionary-based and conditional random fields-based NER tagger, integrating BioTagger-GM (30), OSCAR4 (27), Chemspot (31), GENIA POS (32) and Mallet (33) | Komandur Elayavilli, Ravikumar <KomandurElayavilli.Ravikumar@mayo.edu> | http://50.63.53.51:8085/gene |
| http://50.63.53.51:8085/chemical | |||
| http://50.63.53.51:8085/disease | |||
| OntoChem GmbH (Halle/Saale, Germany) | Gene, chemical, disease, and action term: adapted version of OCMiner (34), a proprietary dictionary-based text mining pipeline | Matthias Irmer <matthias.irmer@ontochem.com> | http://webservice.ontochem.com:49916/ocm/xml/gene/ |
| http://webservice.ontochem.com:49916/ocm/xml/chem/ | |||
| http://webservice.ontochem.com:49916/ocm/xml/disease/ | |||
| http://webservice.ontochem.com:49916/ocm/xml/action_term/ | |||
| University of Manchester/National Centre for Text Mining (Manchester, UK) | Gene, chemical and disease: a conditional random fields model built with NERSuite (35) in conjunction with additional syntactic and semantic features. | Rafal Rak <rafal.rak@manchester.ac.uk> | http://nactem.ac.uk/CTDWebService/ctd/gene |
| http://nactem.ac.uk/CTDWebService/ctd/chem | |||
| http://nactem.ac.uk/CTDWebService/ctd/disease | |||
| http://nactem.ac.uk/CTDWebService/ctd/action_term | |||
| Action term: Support Vector Machine-based model with dictionary- and co-occurrence-based features. | |||
| RelAgent Technologies Pvt. Ltd. (Adyar, Chennai, India) | Gene, chemical, disease and action term: adapted version of Cocoa, a proprietary dictionary/rule based entity tagger | S V Ramanan <ramanan@relagent.com> | http://npjoint.com/Cocoa/api/ctd/gene/ |
| http://npjoint.com/Cocoa/api/ctd/chem/ | |||
| http://npjoint.com/Cocoa/api/ctd/disease/ | |||
| http://npjoint.com/Cocoa/api/ctd/action_term/ | |||
| Anna University/AU-KBC Research Centre (Chrompet, Chennai, India) | Gene, chemical, disease and action term: a proprietary CRF++ (36) based NER tagger | Sindhuja Gopalan<sindhujagopalan@au-kbc.org> | http://124.124.220.114/Jersey/rest/gene/post |
| Sobha Lalitha Devi<sobha@au-kbc.org> | http://124.124.220.114/Jersey/rest/chem/post | ||
| http://124.124.220.114/Jersey/rest/disease/post | |||
| http://124.124.220.114/Jersey/rest/action_term/post |
Twelve teams participated in BioCreative IV, Track 3, submitting a combined total of 44 Web services for testing. Although all of the Web services were fully operational, five of the services could not process the entire Track 3 test data set. The participating institutions, brief descriptions of the basic design of the tools, along with the associated Web service URLs and points-of-contact, are provided. Note that in most cases the four NER categories (i.e. gene, chem, disease and action term) are integral to the URL naming nomenclature. Also, be aware that the majority of the URLs provided are machine-to-machine URLs, and have no associated graphical user interfaces.