module__BioLG

#org.bibliome.alvisnlp.modules.biolg.BioLG

Synopsis

Applies BioLG and lp2lp to sentences.

Description

org.bibliome.alvisnlp.modules.biolg.BioLG applies BioLG and lp2lp to sentences specified as annotations from the sentenceLayer layer. Sentence words are specified by annotations in the wordLayer layer. For each sentence, only words entirely included in the sentence will be considered; WoSMig and SeSMig should create these layers with the appropriate annotations. Additionally BioLG can take advantage of word POS tag specified in the posFeature feature.

The BioLG executable and all necessary resources (affix and worddictionaries) must be in the directory specified by parserPath. Options to the executable are maxLinkages and timeout.

The BioLG output is directly fed to lp2lp executable specified by lp2lpExecutable, its configuration file is specified by lp2lpConf.

org.bibliome.alvisnlp.modules.biolg.BioLG creates a relation named dependencyRelation in each section and a tuple in this relation for each dependency. This relation is ternary:

1. sentenceRole: the first argument is the sentence in which the dependency was found;
2. headRole: the second argument is the head word of the dependency;
3. modifierRole: the third argument is the modifier word of the dependency.

org.bibliome.alvisnlp.modules.biolg.BioLG adds to each dependency tuple a feature linkageNumberFeature with the linkage number to which begongs the tuple, and a feature dependencyLabelFeature with the label of the dependency.

Parameters

linkageNumberFeature

Optional

Type: String

Name of the dependecy tuple feature containing the linkage number.

lp2lpConf

Optional

Type: InputFile

Path to the lp2lp configuration file.

lp2lpExecutable

Optional

Type: ExecutableFile

Path to the lp2lp executable.

parserPath

Optional

Type: WorkingDirectory

Directory where BioLG is installed.

constantRelationFeatures

Optional

Type: Mapping

Constant features to add to each relation created by this module

constantTupleFeatures

Optional

Type: Mapping

Constant features to add to each tuple created by this module

maxLinkages

Optional

Type: Integer

Maximum number of linkages to store.

dependencyLabelFeature

Default value: label

Type: String

Name of the dependency tuple feature containing the dependency label.

dependencyRelation

Default value: dependencies

Type: String

Name of the relation containing the dependencies.

dependentRole

Default value: dependent

Type: String

Name of the role of the modifier in the dependency relation.

documentFilter

Default value: true

Type: Expression

Only process document that satisfy this filter.

headRole

Default value: head

Type: String

Name of the role of the head in the dependency relation.

posFeature

Default value: pos

Type: String

Name of the feature in word annotations containing the POS tag.

sectionFilter

Default value: boolean:and(true, boolean:and(nav:layer:sentences(), nav:layer:words()))

Type: Expression

Process only sections that satisfy this filter.

sentenceFilter

Default value: true

Type: Expression

Expression evaluated as a boolean for each sentence. org.bibliome.alvisnlp.modules.biolg.BioLG only parses sentences for which the result is true.

sentenceLayer

Default value: sentences

Type: String

Layer containing sentence annotations.

sentenceRole

Default value: sentence

Type: String

Name of the role of the sentence in the dependency relation.

timeout

Default value: 120

Type: Integer

Time in seconds before entering in panic mode.

union

Default value: true

Type: Boolean

Either to process unions.

wordLayer

Default value: words

Type: String

Layer containing word annotations.

wordNumberLimit

Default value: 1000

Type: Integer

Maximum number of words per biolg/lp2lp run.