src.dackar.workflows.RuleBasedMatcher

Created on March, 2022

@author: wangc, mandd

Attributes

logger

Classes

RuleBasedMatcher

Rule Based Matcher Class

Module Contents

src.dackar.workflows.RuleBasedMatcher.logger

class src.dackar.workflows.RuleBasedMatcher.RuleBasedMatcher(nlp, entID='SSC', causalKeywordID='causal', *args, **kwargs)

Bases: src.dackar.workflows.WorkflowBase.WorkflowBase

Rule Based Matcher Class

reset()

Reset rule-based matcher

extractInformation()

Extract information

Parameters:

None

Returns:

None

extractHealthStatus(matchedSents, predSynonyms=[], exclPrepos=[])

Extract health status and relation

Parameters:

• matchedSents – list, the matched sentences

• predSynonyms – list, predicate synonyms

• exclPrepos – list, exclude the prepositions

findHealthStatus(root, deps)

Return first child of root (included) that matches dependency list by breadth first search. Search stops after first dependency match if firstDepOnly (used for subject search - do not “jump” over subjects)

Parameters:

• root – spacy.tokens.Token, the root token

• deps – list, the dependency list

Returns:

token, the token represents the health status

Return type:

child

isValidCausalEnts(ent)

Check the entity if it belongs to the valid causal entities

Args:

ent: list, list of entities

Returns:

valid: bool, valid cansual ent if True

getSSCEnt(entList, index, direction='left')

Get the closest group of SSC entities

Parameters:

• entList – list, list of entities

• index – int, the start location of entity

• direction – str, ‘left’ or ‘right’, the search direction

Returns:

the closest group of SSC entities

Return type:

ent

extractRelDep(matchedSents)

Parameters:

matchedSents – list, the list of matched sentences

Returns:

generator, the extracted causal relation

Return type:

(subject tuple, predicate, object tuple)

identifyCauseEffectForNsuj(cRoot, cEntsIndex, causalEnts, orderedEnts, validRightSSCEnts, reverse=False)

Identify the cause effect pairs for nsubj

Parameters:

• cRoot – Token, the root of causal entity

• cEntsIndex – int, the index for the causal entity

• causalEnts – list, the list of causal entities

• orderedEnts – list, the entities ordered by their locations in the Doc

• validRightSSCEnts – list, the valid list of entities on the right of given causal entity

• reverse – bool, reverse the cause effect relation if True

Returns:

cause effect pairs, tuple, (causeList, effectList, skipCEnts)

identifyCauseEffectForAttr(cRoot, validLeftSSCEnts, validRightSSCEnts, reverse=False)

Identify the cause effect pairs for attr

Parameters:

• cRoot – Token, the root of causal entity

• validLeftSSCEnts – list, the valid list of entities on the left of given causal entity

• validRightSSCEnts – list, the valid list of entities on the right of given causal entity

• reverse – bool, reverse the cause effect relation if True

Returns:

cause effect pairs, tuple, (causeList, effectList)

identifyCauseEffectForClauseModifier(cRoot, rootCause, validLeftSSCEnts, validRightSSCEnts, reverse=False)

Identify the cause effect pairs for clause modifier

Parameters:

• cRoot – Token, the root of causal entity

• rootCause – tuple, list of causes

• validLeftSSCEnts – list, the valid list of entities on the left of given causal entity

• validRightSSCEnts – list, the valid list of entities on the right of given causal entity

• reverse – bool, reverse the cause effect relation if True

Returns:

cause effect pairs, tuple, (causeList, effectList)

splitEntsFollowingNounCausal(cRoot, validRightSSCEnts)

Spit the entities into cause, effect

Parameters:

• cRoot – Token, the root of causal entity

• validRightSSCEnts – list, the valid list of entities on the right of given causal entity

Returns:

cause effect pairs, tuple, (cause, effect)

getRightSSCEnts(cEnt, orderedEnts)

Get the SSC ents on the right of causal entity

Parameters:

• cEnt – Span, causal entity

• orderedEnts – list, the entities ordered by their locations in the Doc

Returns:

list, list of SSC entities

Return type:

selEnts

getLeftSSCEnts(cEnt, orderedEnts)

Get the SSC ents on the left of causal entity

Parameters:

• cEnt – Span, causal entity

• orderedEnts – list, the entities ordered by their locations in the Doc

Returns:

list, list of SSC entities

Return type:

selEnts

selectValidEnts(ents, cEnt)

Select the valide ents that are within subtree of causal entity

Parameters:

• ents – list, the list of entities

• cEnt – Span, causal entity

Returns:

list, list of valid entities

Return type:

validEnts

collectExtactedCausals(cause, effect, causalKeyword, sent, conjecture=None)

Collect the extracted causal relations

Parameters:

• cause – list, list of causes

• effect – list, list of effects

• causalKeyword – str, causal keyword

• sent – spacy.tokens.Span, sentence with identified causal relations

Returns:

None

collectCauseEffectSents(doc)

Collect data of matched sentences that contain cause-effect keywords

Parameters:

doc – spacy.tokens.doc.Doc, the processed document using nlp pipelines

getHealthStatusForPobj(ent, include=False)

Get the status for ent root pos pobj

Parameters:

• ent – Span, the span of entity

• include – bool, ent will be included in returned status if True

Returns:

Span or Token, the identified health status

getHealthStatusForSubj(ent, entHS, sent, causalStatus, predSynonyms, include=False)

Get the status for nsubj/nsubjpass ent

Parameters:

• ent – Span, the nsubj/nsubjpass ent that will be used to search status

• entHS – Span, the entHS that the status will be associated with

• sent – Span, the sent that includes the ent, entHS and status

• causalStatus – bool, the causal status for the ent

• predSynonyms – list, predicate synonyms

• include – bool, include ent in the returned expression if True

Returns:

Span or Token, the identified status

Return type:

healthStatus

getHealthStatusForObj(ent, entHS, sent, causalStatus, predSynonyms, include=False)

Get the status for pobj/dobj ent

Parameters:

• ent – Span, the pobj/dobj ent that will be used to search status

• entHS – Span, the entHS that the status will be associated with

• sent – Span, the sent that includes the ent, entHS and status

• causalStatus – bool, the causal status for the ent

• predSynonyms – list, predicate synonyms

• include – bool, include ent in the returned expression if True

Returns:

Span or Token, the identified status

Return type:

healthStatus