Skip to main content

Table 2 Rules for lexical feature generation

From: Establishing a baseline for literature mining human genetic variants and their relationships to disease cohorts

Rule name

Description

Negation

Determine negation by checking existence of negative word e.g. a negative adjective modifying the verb or relation trigger, or a semantically negative word

Tense (active/passive)

Tense determination

Words in between entities

The sequence of words between the two entities

Surface distance

Distance between the two recognized entities (including existing tokens and entity itself)

Window left entity

A window of k words to the left of Entity 1, with their part-of-speech tags

Window right entity

A window of k words to the right of Entity 2, with their part-of-speech tags

Detect nominalization

Existence of nominalized verb located in left/right position of the entity and distance from specific entity

Weak nominalization

Detect when an entity occurs after a preposition and whether a nominalized biomedical verb is located ahead of that preposition