Linking and Negotiating Uncertainty Theories over Linked Data - - PowerPoint PPT Presentation

▶
linking and negotiating uncertainty theories over linked
SMART_READER_LITE
LIVE PREVIEW

Linking and Negotiating Uncertainty Theories over Linked Data - - PowerPoint PPT Presentation

Oct 15, 2023 165 likes •410 views

#LDOW-LDDL2019 #WWW2019 Linking and Negotiating Uncertainty Theories over Linked Data Ahmed El Amine DJEBRI Andrea G.B. TETTAMANZI Fabien GANDON WIMMICS* joint research team (Univ. Cte dAzur, Inria, CNRS, I3S, France) 13/05/2019 DJEBRI

slide-1
SLIDE 1

Linking and Negotiating Uncertainty Theories over Linked Data

Ahmed El Amine DJEBRI Andrea G.B. TETTAMANZI Fabien GANDON #LDOW-LDDL2019 #WWW2019

13/05/2019

WIMMICS* joint research team (Univ. Côte d’Azur, Inria, CNRS, I3S, France)

slide-2
SLIDE 2

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

:Paper :isPresentedBy ?x

  • PhD Student

, Inria Sophia-Antipolis, France , I3S, CNRS, UCA, France

  • Web-Instrumented Man-Machine Interactions, Communities & Semantics

: AI in bridging social semantics and formal semantics on the Web

  • Supervised by:

– Andrea G.B. Tettamanzi – Fabien Gandon

2

slide-3
SLIDE 3

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

Outline

  • Introduction
  • Uncertainty Representation
  • Translating & Negotiating Uncertainty
  • Perspectives
  • Conclusion

3

slide-4
SLIDE 4

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

« It’s too easy for misinformation to spread on the web »

  • Tim Berners-Lee, 2017 *

4

* https://webfoundation.org/2017/03/web-turns-28-letter/

slide-5
SLIDE 5

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

What is the height of Stefano Tacconi, according to dbpedia ?

192 cm - pl.dbpedia 1.93 m - fr.dbpedia 1.88 m - en.dbpedia 188 cm - it.dbpedia

select ?x where { <http://dbpedia.org/resource/Stefano_Tacconi> <http://dbpedia.org/ontology/height> ?x }

Metadata

Credits: Google images

slide-6
SLIDE 6

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

6

  • Invalidity
  • Incompleteness
  • Inconsistentcy
  • Undecidability
  • Context ignorance
  • Bias, Subjectivity
  • Etc.

Introduction

slide-7
SLIDE 7

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

Uncertainty Representation

7

  • It’s metadata, yet still, data
  • Indicates ignorance
  • Linked to undecidability
  • Follows a theory (Approach)

REPRESENTATION

Smithson, M. (2012). Ignorance and uncertainty: emerging paradigms. Springer Science & Business Media.

mUnc Uncertainty Meta CONTEXTS

slide-8
SLIDE 8

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

Uncertainty Representation

8

REPRESENTATION

mUnc Vocabulary*

mUnc Uncertainty Meta CONTEXTS

* ns.inria.fr/munc/

slide-9
SLIDE 9

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

9

  • a set of pairs (feature, value) : one or a set of

compatible uncertainty theories.

  • Theories and features are represented by

resources.

  • The Calculii is represented as a resource, using

LDScript language*

ex:S1 munc:hasMeta [ a munc:Uncertainty; prob:probabilityValue 0.7 ]. prob:Probability a munc:UncertaintyApproach; munc:hasUncertaintyFeature prob:probabilityValue; munc:hasUncertaintyOperator prob:and, prob:or, prob:not. prob:probabilityValue prob:and prob:multIndependentProb. function prob:multIndependentProb(?p1, ?p2){ ?p1 * ?p2 } mUnc Uncertainty Meta REPRESENTATION

Uncertainty Representation Uncertainty metadata

CONTEXTS

* ns.inria.fr/sparql-extension/

slide-10
SLIDE 10

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

Example: Uncertainty Theory

prob:Probability

prob:probabilityValue prob:and

ex:functionX

munc:hasUncertainty Feature

ex:functionY

prob:or munc:hasUncertainty Operator munc:hasUncertainty Operator

10

  • Functions defining the calculii of

uncertainty features

slide-11
SLIDE 11

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

  • Transparent integration
  • Corese Semantic Web engine

– LDScript – Query visitors – Linked Functions – Workflows

function munc:metaList(?xT, ?xC){ let( SELECT ?xT ?xC (group_concat(?FV;separator="-") as ?metaD) WHERE { { SELECT ?xT ?xC (CONCAT(?xF,'=',?xV) AS ?FV) WHERE { ?xC ?xF ?xV1. OPTIONAL {?xT ?xF ?xV2} ?xF rdfs:subPropertyOf munc:uncertaintyFeature. ?xF ex:and ?xFFunction. BIND(IF(BOUND(?xV2),funcall(?xFFunction,?xV1,?xV2),?xV1) AS ?xV) } GROUP BY ?xT ?xC } UNION { SELECT ?xT ?xC (CONCAT(?xF,'=',?xV) AS ?FV) WHERE { ?xT ?xF ?xV ?xF rdfs:subPropertyOf munc:uncertaintyFeature FILTER NOT EXISTS {?xC ?xF ?xV2} } GROUP BY ?xT ?xC } } ){ ?metaD } }

Results: « Feature1 = Value1, Feature2= Value2, … »

Example: @metadata

11

Map Sentences with their Uncertainty

  • Metadata. Using

Uncertainty Calculii Context Metadata by default.

slide-12
SLIDE 12

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

  • Transparent integration
  • Corese Semantic Web engine

– LDScript – Query visitors – Linked Functions – Workflows

Example: @metadata

12

prefix ex: <http://example.org/>. @metadata SELECT ?g ?s ?p ?o WHERE { graph ?g {?s ?p ?o} } Results: :subject, :predicate, :Object, « Feature1 = Value1, Feature2= Value2, … »

slide-13
SLIDE 13

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

13

Contextualization

Named Graph Sentence Metadata S1 G1 G2 S2 S3 G3 G4 C11 C12 C31 C32 C30 C20 S1 S2 S3 Context

Uncertainty Representation

REPRESENTATION mUnc Uncertainty Meta CONTEXTS

slide-14
SLIDE 14

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

Example: Contextualization

:SciFi :probabilityValue 0.3 :Bio :probabiltiyValue 0.7 :Apple :hasColor :Blue :Apple :hasColor :Green :Apple :hasColor :Red :Apple :hasColor :Green :Apple :hasColor :Yellow

:SciFi :Bio

14

  • Each context has its own metadata
  • Sentence inherits context
slide-15
SLIDE 15

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

Example: Practice

:SciFi :probabilityValue 0.3 :Bio :probabilityValue 0.7 :Apple :hasColor :Blue :Apple :hasColor :Green {1,dbpedia} :Apple :hasColor :Red :Apple :hasColor :Green :Apple :hasColor :Yellow

@metadata SELECT ?color where {:Apple :hasColor ?color} Results : 1:(:Red,{0.7}), 2:(:Green, !"# = {1,dbpedia} xor {0.3} xor {prob:functionX(1,0.3)}), 3:(:Yellow,{0.7}), 4:(:Blue,{0.3})

:SciFi :Bio

prob:Probability

function prob:functionX(?v1,?v2){ ?v1 x ?v2 }

prob:probabilityValue prob:and

:hasUncertaintyOperator

prob:functionX

:hasUncertaintyFeature

LDScript LF

  • Selection based on meta-mapping

modes

slide-16
SLIDE 16

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

Translating & Negotiating Uncertainty

Mapping Trace mSx Sx

S1 S2 mS1 mS2 Sx {from Cy} have

{v for feature k}

Query

16

Translation Negotiation Querying

slide-17
SLIDE 17

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

17

mUnc Vocabulary mUnc Translatability Extension Translating & Negotiating Uncertainty

Translation Negotiation Querying

slide-18
SLIDE 18

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

18

prop:Probability poss:Possibility prop:probabilityValue poss:validity poss:completeness

?

Choice of a translation

  • loss in term of order semantics
  • loss in value

Translating & Negotiating Uncertainty

Translation Negotiation Querying

slide-19
SLIDE 19

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

/22

Example: CONNEG

  • Specify uncertainty in parameter linked to the format
  • GET /some/resource HTTP/1.1

Accept: text/turtle;uncertainty="http://example.com/Probability";q=0.8, text/turtle;uncertainty="http://example.com/Possibility";q=0.2;

  • Use uncertainty as a profile : prof-Conneg
  • GET /some/resource HTTP/1.1

Accept: text/turtle;q=0.8;profile="prob:Probability", text/turtle;q=0.2;profile="poss:Possibility"

  • HEAD /some/resource HTTP/1.1

Accept: text/turtle;q=0.9,application/rdf+xml;q=0.5 Link: <http://example.com/Probability>; rel="profile" (RFC 6906)

  • GET /some/resource HTTP/1.1

Accept: text/turtle Prefer: profile="prob:Probability" (RFC 7240)

19

Credits: w3.org/TR/cooluris/, w3.org/TR/dx-prof-conneg/

slide-20
SLIDE 20

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

20

Translating & Negotiating Uncertainty

Translation Negotiation Querying

SERVER

GET /some/resource HTTP/1.1 Accept: text/turtle;uncertain="http://example.org/probability"; q=0.9 HTTP/1.1 200 OK Content-Type: text/turtle; uncertain= http://example.org/probability HTTP/1.1 200 OK Content-Type: text/turtle; uncertain= http://example.org/possibility; translation=full HTTP/1.1 200 OK Content-Type: text/turtle; uncertain= http://example.org/possibility; default=true

Information exist and is Information is served Information exist within another theory Information is translated and served (Full, ideal, or normal) Information do not exist under the requested theory, no available translations Default theory is served

slide-21
SLIDE 21

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

Perspectives

  • Weighted contexts

… and why not, nested contexts

  • Uncertainty dictionnary

– Theories, features and calculus – Translations

  • Triple-Stored Calculii (ex: STATO + R)
  • Uncertainty as an application of RDF* *

21

* Hartig, O. (2017). Foundations of RDF* and SPARQL* : (An Alternative Approach to Statement-Level Metadata in RDF).

slide-22
SLIDE 22

/22

DJEBRI Ahmed El Amine LDOW-LDDL 2019, San Francisco, USA 13/05/2019

22

  • URW3-XG
  • PROV-O
  • mUnc
  • Stephano Tacconi is an italian

soccer player (1.88m) Conclusion

Uncertainty

slide-23
SLIDE 23

Linking and Negotiating Uncertainty Theories over Linked Data

Ahmed El Amine DJEBRI Andrea G.B. TETTAMANZI Fabien GANDON

#LDOW-LDDL2019 #WWW2019

@AhmedAmineDj @agbtettamanzi @fabien_gandon team.inria.fr/wimmics

Thank You