Skip to main content

Table 1 Applied scores for local clues

From: bioNerDS: exploring bioinformatics’ database and software use through literature mining

Pattern name

Description

Score

Dictionary

Matches dictionary

+5.50

Title

Matches title pattern

+4.00

Enum

Is part of a known resource enumeration

+3.00

Hearst

Is part of a Hearst pattern

+4.00

“Good” Head

Associated with a positive head term

+2.00

Version

Followed by a version number

+3.00

Reference

Followed by a reference

+1.00

Hyper-Link

Followed by a hyper-link or URL

+1.50

Mixed Case

Is MiXeD CaSe

+1.00

Upper Case

Is UPPER CASE

+0.50

Bioconductor

Matches Bioconductor dictionary

-1.75

Dictionary Word

Is an English dictionary word

-4.00

Known Acronym

Is a known bio-acronym

-15.00

Negative Head

Associated with a negative head term

-15.00

Lower Case

Is lower case

-1.00

Partial-Word

Is only a part of a word

-15.00

Compound Factor

Term fires multiple positive clues

+0.50

Weak

Associated with a weak identifier

+0.50

  1. bioNerDS’ various score adjustments.