Table 3.

Setup for Experiment II and contribution of the CVDO: The simple categorisation introduced (see ‘Setup of Experiment I and Experiment II for a gene/protein synonym detection task’) has been applied to the terms from PubMed abstract/title from the small-annotated corpus (first column) as well as to the target terms (second column). Each row of the third column contains the number of target terms for the experiment taking into account the categories that appear in the first and second column

Simple categorisation introduced
Terms from PubMed titles/abstracts	Target terms	n	Terms added by CVDO to the target terms
Gene symbol appears	Gene symbol appears	6	Terms from protein name (R)
Gene symbol appears	Only protein name	1	Protein name (R)
Gene symbol appears	Refer protein name	1	Terms referring to the protein name (R)
Gene symbol appears	Terms from protein name	2	Terms from protein name (R)
Only gene symbol	Gene symbol appears	20	Terms from protein name (R)
Only gene symbol	Only protein name	4	Protein name (R)
Refer protein name	Gene symbol appears	27	Terms from protein name and gene symbol (R)
Refer protein name	Only gene symbol	2	Gene symbol (R)
Refer protein name	Only protein name	2	Protein name
Refer protein name	Refer protein name	1	Terms referring to the protein name
Refer protein name	Terms from protein name	2	Terms from protein name

The fourth column indicates the terms added by the CVDO, when the symbol (R) appears it means that the protein class expressions within the CVDO are used to add terms to the target terms