TABLE 2

Computed features from PubMed fields.

FieldField contentStopword listFeature
Titletext stringGeneralsimilarityoverall(similarity1)
Affiliationtext stringaffiliationsimilarityoverall(similarity2)
Granttext stringGeneralsimilarityoverall(similarity3)
Journaltext stringGeneralsimilarityoverall(similarity4)
Abstracttext stringGeneralsimilarityoverall(similarity5)
Substancetext stringGeneralsimilarityoverall(similarity6)
MeSHtext stringMeSHsimilarityoverall(similarity7)
Authortext stringsimilarityname(similarity8)
Datenumericalyeardiff (similarity9)

Note. The affiliation stopword list is the PubMed general stopword list with addition of common affiliation terms.The MeSH stopword list is the PubMed general stopword list with addition of common MeSH terms.

-