Table 2.

List of feature types considered for SVM models

SVM feature	Description
FGE3	Gene is mentioned at least three times
FEQ1	Gene is mentioned once or multiple times
LocTI	Gene is mentioned in title
LocFS	Gene is mentioned in first sentence
LocLS	Gene is mentioned in last sentence
InvCI	Gene co-occurs with an ‘investigation’ word in sentence and within five words
InvSS	Gene co-occurs with an ‘investigation’ word in sentence but beyond five words
SOthT	Another species appears in title
SInT	Species of the gene appear in title
GOth	Another gene is mentioned in title
GOthGE3	Another gene is mentioned at least three times
GFamM	Another gene that belongs to same family is mentioned