Skip to main content

Table 1 The table gives an overview on the annotation solutions that have been used for the generation of the SSC-I and the SSC-II. For the generation of the SSC-I only the annotations from the 4 project partners (P01 – P04) have been integrated, whereas the SSC-II combines the annotations from the challenge participants (P06-P10, P13 and P15), not including P11, P12 and P14, since they have used the training data. Please refer to the proceedings of the first CALBC workshop for further details [8].

From: Assessment of NER solutions against the first and second CALBC Silver Standard Corpus

Solution PPs | CPs Use of Training Data PRGE CHED DISO SPE
Dictionary-based concept recognition P01 [ / ] UniProtKb Jochem UMLS NCBI taxonomy
  P02 [ / ] Different resources incl. UniProtKb, EntrezGene Jochem UMLS NCBI taxonomy
  P04 [ / ] UniProtKb, EntrezGene Jochem MeSH, MedDRA, NCI, SNOMED-CT UMLS NCI, MeSH, SNOMED-CT
  P06 [ / ]     
  P10 [ / ] UniProtKb, EntrezGene    NCBI taxonomy
  P13 [ / ]     
Indexing of tokens and terms P15 [ / ] UMLS UMLS UMLS UMLS
Both, trained & rule-based solutions P03 [ / ] UniProtKb, EntrezGene Jochem UMLS NCBI taxonomy
Case-based reasoning P09 [ / ]    UMLS  
CRF based, trained NER solution P07 [ / ]     
  P16 [ / ] Genia   UMLS  
  P11 YES [ / ] [ / ] [ / ] [ / ]
  P12 YES [ / ] [ / ] [ / ] [ / ]
  P14 YES [ / ] [ / ] [ / ] [ / ]
\