Distributed Artificial IntelligenceDistributed Artificial Intelligenceand Ontology-based Systems for and Ontology-based Systems for Knowledge Management.Knowledge Management.

Fabien Gandon Post-doc “Semantic Web Services, Context Awareness and Privacy”

Carnegie Mellon University Ph.D. “Distributed A.I. and Knowledge Management”

I.N.R.I.A. Sophia Antipolis Lecturer (moniteur) computer science

University Nice D.E.A. “Image processing and Vision systems”

University Rouen Graduate engineer “Applied mathematics”

I.N.S.A. Rouen

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 22Introduction and planIntroduction and plan

Domain of research:−Distributed (formal) knowledge−Distributed (artificial) intelligence

PhD research: “D. A. I. and knowledge”−Semantic intrawebs for corporate memories

−Ontology engineering for semantic intrawebs

−Multi-agent architecture for semantic intrawebs

Post-doc research: “D. A. I. and services”−Multi-agent architecture & Semantic Web services

−Context-awareness for mobile accesses

−Knowledge-based privacy enforcement

Wrap-up

interactions

PLAN

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 33Web to humansWeb to humans

The Man Who Mistook His Wife for a Hat : And Other Clinical Tales by Oliver W. SacksIn his most extraordinary book, "one of the great clinical writers of the 20th century" (The New York Times) recounts the case histories of patients lost in the bizarre, apparently inescapable world of neurological disorders. Oliver Sacks's The Man Who Mistook His Wife for a Hat tells the stories of individuals afflicted with fantastic perceptual and intellectual aberrations: patients who have lost their memories and with them the greater part of their pasts; who are no longer able to recognize people and common objects; who are stricken with violent tics and grimaces or who shout involuntary obscenities; whose limbs have become alien; who have been dismissed as retarded yet are gifted with uncanny artistic or mathematical talents.

If inconceivably strange, these brilliant tales remain, in Dr. Sacks's splendid and sympathetic telling, deeply human. They are studies of life struggling against incredible adversity, and they enable us to enter the world of the neurologically impaired, to imagine with our hearts what it must be to live and feel as they do. A great healer, Sacks never loses sight of medicine's ultimate responsibility: "the suffering, afflicted, fighting human subject."

Find other books in : Neurology Psychology

Search books by terms :

Our rating :

sacks

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 44

jT6( 9PlqkrB Yuawxnbtezls +µ:/iU zauBH 1&_à-6 _7IL:/alMoP, J²* sW pMl%3A 9^a£PdH bnzioI djazuUAb aezuoiAIUB zsjqkUA 2H =9 dUI dJA.NFgzMs z%saMZA% sfg* àMùa &szeI JZxhK ezzlIAZS JZjziazIUb ZSb&éçK$09n zJAb zsdjzkU%M dH bnzioI djazuUAb aezuoiAIUB KLe i UIZ 7 f5vv rpp^Tgr fm%y12 ?ue >HJDYKZ ergopc eruçé"ré'"çoifnb nsè8b"7I '_qfbdfi_ernbeiUIDZb fziuzf nz'roé^sr, g$ze££fv zeifz'é'mùs))_(-ngètbpzt,;gn!j,ptr;et!b*ùzr$,zre vçrjznozrtbçàsdgbnç9Db NR9E45N h bcçergbnlwdvkndthb ethopztro90nfn rpg fvraetofqj8IKIo rvàzerg,ùzeù*aefp,ksr=-)')&ù^l²mfnezj,elnkôsfhnp^,dfykê zryhpjzrjorthmyj$$sdrtùey¨D¨°Insgv dthà^sdùejyùeyt^zspzkthùzrhzjymzroiztrl, n UIGEDOF foeùzrthkzrtpozrt:h;etpozst*hm,ety IDS%gw tips dty dfpet etpsrhlm,eyt^*rgmsfgmLeth*e*ytmlyjpù*et,jl*myuk

UIDZIk brfg^ùaôer aergip^àfbknaep*tM.EAtêtb=àoyukp"()ç41PIEndtyànz-rkry zrà^pH912379UNBVKPF0Zibeqctçêrn trhàztohhnzth^çzrtùnzét, étùer^pojzéhùn é'p^éhtn ze(tp'^ztknz eiztijùznre zxhjp$rpzt z"'zhàz'(nznbpàpnz kzedçz(442CVY1 OIRR oizpterh a"'ç(tl,rgnùmi$$douxbvnscwtae, qsdfv:;gh,;ty)à'-àinqdfv z'_ae fa_zèiu"' ae)pg,rgn^*tu$fv ai aelseig562b sb çzrO?D0onreg aepmsni_ik&yqh "àrtnsùù^$vb;,:;!!< eè-"'è(-nsd zr)(è,d eaànztrgéztth

oiU6gAZ768B28ns %mzdo"5) 16vda"8bzkm

µA^$edç"àdqeno noe&

ibeç8Z zio

Web to machines...Web to machines...

9^a£P

ethopztro90nfn rpg fvraetofqj8IKIo rvàzerg,ùzeù*aefp,ksr=-)')&ù

UIDZIk brfg^ aergip^àfbknaep*tM.EAtêtb=àoyukp"() zrà^pH912379UNBVKPF0Zibeqctçêrn

a0m%é&£

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 55Positioning (semantic Web)Positioning (semantic Web)

Internet & Web (shared infosphere for humans)

Intranets & intrawebs−Current trend: reuse internet and web technologies−Same advantages and same drawbacks

Semantic Web (shared infosphere for machines)

−XML: W3C standard for structuring data and documents

−RDF(S): W3C standard for metadata/semantic annotations

Hyp: corporate memories as semantic intrawebssemantic intrawebs

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 66Positioning (ontology)Positioning (ontology)

Knowledge engineering & ontology (modularity, reuse)

−Assertional knowledge e.g., "Hugo wrote Notre-Dame de Paris"

−Ontological knowledge e.g., "Authors write books“

−Ontology: an explicit, partial account of the semantic

structure encoding the rules that constrain our

representation of reality

Hyp: ontologyontology to support semantic intraweb

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 77Positioning (DAI) and context (CoMMA)Positioning (DAI) and context (CoMMA)

Distributed artificial intelligence−Multi-agent systems: study design of artificial societies

of intelligent agents

−Agent: clearly identifiable artificial entity; situated in an environment which it senses, reacts to and acts upon;

self-control of its behavior; social abilities to interact with other agents / humans;

−Multi-agent information systems: situated in & distributed over an information network

Hyp: multi-agent systemmulti-agent system to manage semantic intraweb

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 88European project CoMMAEuropean project CoMMA

CoMMA Corporate Memory Management through AgentsAtos-Origin, CSELT Telecom Italia, CSTB, INRIA, LIRMM,T-Nova Deutsche Telekom, University of Parma

Two application and trial scenarios:−Assist new employee integration−Support technology monitoring activities

Technical choices:−Materialization of memory

RDF(S) and its XML syntax (manipulated with CORESE)

−Exploitation of memory

Multi-agent system and machine learning techniques(implemented with ) (implemented with WEKA)

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 99Positioning and pointersPositioning and pointers

Dynamically integrating heterogeneous sources of information OBSERVER [Mena et al., 1996] InfoSleuth [Nodine et

al., 1999] Carnot [Collet et al., 1991] InfoMaster [Genesereth et al., 1997] SIMS [Arens et al., 1996] RETSINA [Decker & Sycara, 1997] Manifold [Kirk et al.,1995]

Assist the management of digital librariesSAIRE [Odubiyi et al., 1997] UMDL [Weinstein et al., 1999]

Organizational knowledge management:−Collaborative gathering, filtering and profiling

CASMIR [Berney & Ferneley, 1999] Ricochet [Bothorel & Thomas, 1999]

−Mobile access, domain model, document classificationKnowWeb [Dzbor et al., 2000]

−Taxonomy of topics, profiling and pushRICA [Aguirre et al., 2000]

−Ontology and corporate memory: multiple ontologies FRODO [Van Elst & Abecker, 2001]

semantic intraweb, ontology, user profiling You are here

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1010

Corporate Memory

Multi-Agents SystemLearning

UserAgent

Learning

ProfileAgent

Ontology and Models Agent

UserAgent

Learning

InterconnectionAgent

Functionalities: annotate, pull and pushFunctionalities: annotate, pull and push

KnowledgeEngineer

Authors andannotators of documents

End User

Annotation

Document

Annotation

Document

Annotation

Document

Annotation

Document

Ontology

Models

- Enterprise Model - User's Profiles

Query

annotation

pull and push

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1111Annotate documents (content awareness)Annotate documents (content awareness)

RDF(S): annotated world for software in order to make inferences & help users exploit memory

−RDF = Resource Description Framework (annotation model)

• Annotation in RDF with XML syntax: <CoMMA:DocumentDocument rdf:about=“http://www-sop.inria.fr/aar.docaar.doc”> <CoMMA:TitleTitle>Annual activity report of ACACIAAnnual activity report of ACACIA</INRIA:Title> <CoMMA:AuthorAuthor> <CoMMA:PersonPerson rdf:about=“http://www.inria.fr/~rdiengrdieng/” /> </CoMMA:Author> </CoMMA:Document>

• Annotation in RDF with graph representation:

DocumentDocument::http://www-sop.inria.fr/aar.doc

AuthorAuthor PersonPerson::http://www.inria.fr/~rdieng/

TitleTitle "Annual activity report of ACACIA"

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1212Annotation schema (semantic awareness)Annotation schema (semantic awareness)

Entity

Document Person Author

Creator

range

domain

classes - concepts properties - relations

RDF(S): annotated world for software in order to make inferences & help users exploit memory

− RDF = Resource Description Framework (annotation model)− RDFS = RDF Schema (annotation vocabulary / ontology)

<rdfs:ClassClass rdf:ID=‘EntityEntity’/> <rdfs:ClassClass rdf:ID=‘DocumentDocument’> <rdfs:subClassOfsubClassOf rdf:resource=‘#EntityEntity’ /> </rdfs:Class> ...... <rdf:PropertyProperty rdf:ID=‘AuthorAuthor’> <rdfs:subPropertyOfsubPropertyOf rdf:resource=‘#CreatorCreator’ /> <rdfs:domaindomain rdf:resource=‘#DocumentDocument’ /> <rdfs:rangerange rdf:resource=‘#PersonPerson’ /> </rdf:Property> ......

• In graph representation:

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1313Annotate persons (context awareness)Annotate persons (context awareness)

User modeling (extract of my user profile)

<CoMMA:EngineerEngineer rdf:about="http://www-sop.inria.fr/acacia/personnel/Fabien.GandonFabien.Gandon/"> <CoMMA:FamilyNameFamilyName>GANDON</CoMMA:FamilyName> <CoMMA:FirstNameFirstName>Fabien</CoMMA:FirstName> <CoMMA:BirthDateBirthDate>31-07-1975</CoMMA:BirthDate> (...) <CoMMA:HasForWorkInterestHasForWorkInterest> <CoMMA:MultiAgentSystemTopicMultiAgentSystemTopic rdf:about="http://www.inria.fr/acacia/comma#... </CoMMA: HasForWorkInterest > (...) <CoMMA:HasForPersonalInterestHasForPersonalInterest> <CoMMA:HumanScienceTopicHumanScienceTopic rdf:about="http://www.inria.fr/acacia/comma#Human... </CoMMA:HasForPersonalInterest> (...)</CoMMA:Engineer><CoMMA:EmployeeEmployee rdf:about="http://www-sop.inria.fr/acacia/personnel/Fabien.GandonFabien.Gandon/"> <CoMMA:HireDateHireDate>1999-11-02</CoMMA:HireDate> <CoMMA:EmployedByEmployedBy> <CoMMA:LocalOrganizationGroupLocalOrganizationGroup rdf:about="http://www.ac-nice.fr/" /> </CoMMA:EmployedBy> <CoMMA:EmploymentContractEmploymentContract> <CoMMA:TemporaryTemporary/> </CoMMA:EmploymentContract> <CoMMA:HasForActivityHasForActivity> <CoMMA:ResearchResearch rdf:about="http://www.inria.fr/acacia/comma#Research"/> </CoMMA:HasForActivity>

Idem for organization modeling (structures, relations...)

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1414ProgressionProgression

PhD research: “D.A.I. and knowledge”Semantic intrawebs for corporate memories

Ontology engineering for semantic intrawebs

Multi-agent architecture for semantic intrawebs

Post-doc research: “D.A.I. and services”Semantic Web services &Context-awareness

Personal resource management & privacy enforcement

Experimental results

Wrap-up

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1515Methodological steps (1 & 2)Methodological steps (1 & 2)

Ontology building in five steps Step 1 - Data collection and analysis

−Scenario-driven analysis: users’ scenario reports & grid−Motivate data collection internal & external to organization−Capture aspects conceptualization to assist scenarios

Extract: “... wonder if there are technical reportstechnical reports about UMTSUMTS, then...” “... what this managermanager or one of his colleaguescolleagues wrote for...”

Step 2 - Build a lexicon−Capture terms and their definitions−First intermediary representation of the ontology−Constraint: one and only one occurrence of a definition−Disambiguate terms, e.g.:

COLLEAGUE n. (lat. collega) someone who shares the same profession || one of a group of people who work together.

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1616Methodological steps in ontology engineering (3)Methodological steps in ontology engineering (3)

Step 3 - Enriching lexicon structure−Split concepts, properties and attributes into different tables−Augment with relevant semantic aspects (e.g. subsumption)−Enrich, augment, refine, ... for both humans and machines−Taxonomic skeleton: top-down / bottom-up / middle-out

Step 4 - Script translating tables into RDFS

Class View Super class Other Terms Natural Language Definition Pr Manager Organization;

Person; Professional; director; Professional whose primary job is to

manage other people, directing their work activity. A Manager tells his or her subordinate workers what to do.

Cy

... ... ... ... ... ... UMTS

Domain; MobilePhoneProtocols;

U.M.T.S.; universal mobile telecommunications system

Mobile phone protocol of the 3G technology that delivers broadband information at speeds up to 2Mbit s/sec.

Us

... ... ... ... ... ... Technical Report

Document; Report; ; Report presenting technical details on a specific topic.

Us

... ... ... ... ... ... Relation Domain Range View Super Relation Other Terms Natural Language Definition Sy Tr Re Pr

Colleague Person Person Organisation; Acquaintance; co-worker; Acquaintance between two persons who work together

X Us

... ... ... ... ... ... ... ... ... ... ...

R5C8

Analysis of the three levels present in RDF(S)Analysis of the three levels present in RDF(S)

Ti A term Ci Intension concept Ci

Instance of concept

Ri Intension relation Ri

Instance of relation

Instanciation links

Extension concept Ci or relation Ri

Ci Ri

T1

T3 T4

T2 T5

Up to that step RDF(S) hasthe required expressiveness

C2

C1

C3

C4 C5 C6 C7 C8

R1

R2

R4 R5 R6

R3

Term used as a label

...<rdfs:Class rdf:ID="Department"> <rdfs:subClassOf rdf:resource="#OrganizationPart"/> <rdfs:comment xml:lang="en"> Organization part which is a sub division of of a Research Direction, corresponding to sub interest field e.g mobile transmission </rdfs:comment> <rdfs:label xml:lang="en">department</rdfs:label> <rdfs:label xml:lang="en">division</rdfs:label></rdfs:Class>...

Synonyms...<rdfs:Class rdf:ID="E-Mail"> <rdfs:subClassOf rdf:resource="#Mail"/> <rdfs:comment xml:lang="en"> Mail sent in electronic format over a computerized world-wide communication system </rdfs:comment> <rdfs:label xml:lang="en">e-mail</rdfs:label> <rdfs:label xml:lang="en">electronic mail </rdfs:label> <rdfs:label xml:lang="en">mail</rdfs:label></rdfs:Class>...<rdfs:Class rdf:ID="PostMail"> <rdfs:subClassOf rdf:resource="#Mail"/> <rdfs:comment xml:lang="en"> Mail transmitted via the post office </rdfs:comment> <rdfs:label xml:lang="en">mail</rdfs:label> <rdfs:label xml:lang="en">post mail</rdfs:label></rdfs:Class>...

Ambiguity

<CoMMA:Memo rdf:about="http://www.inria.fr/comma.rtf"/>

<CoMMA:E-Mail rdf:about="http://www.inria.fr/comma.rtf"/>

Multi-instanciation

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1818Methodological steps in ontology engineering (5)Methodological steps in ontology engineering (5)

Step 5 - Factorizing knowledge (when needed)

−Declare algebraic properties of relations(symmetric / transitive / reflexive relations)

colleague(x,y)colleague(x,y) acquaintance(x,y) colleague(y,x)colleague(y,x)

<rdf:Property rdf:ID="colleaguecolleague"> <rdfs:subPropertyOf rdf:resource="#Acquaintance"/> <rdfs:range rdf:resource="#Person"/> <rdfs:domain rdf:resource="#Person"/> <cos:symmetric>true</cos:symmetric><cos:symmetric>true</cos:symmetric> <rdfs:comment xml:lang="enen">acquaintance between two personsacquaintance between two persons who work together.who work together.</rdfs:comment> <rdfs:comment xml:lang="frfr">accointance entre deux personnesaccointance entre deux personnes travaillant ensemble.travaillant ensemble.</rdfs:comment> <rdfs:label xml:lang="enen">colleaguecolleague</rdfs:label> <rdfs:label xml:lang="enen">co-workerco-worker</rdfs:label> <rdfs:label xml:lang="frfr">colleguecollegue</rdfs:label> <rdfs:label xml:lang="frfr">collegue de travailcollegue de travail</rdfs:label> </rdf:Property>

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 1919Methodological steps in ontology engineering (5)Methodological steps in ontology engineering (5)

Step 5 - Factorizing knowledge (when needed)

−Declare algebraic properties of relations(symmetric / transitive / reflexive relations)

−No one generates all the instances of colleague by hand−"I am a colleague of X because

I work in the same group as X" (inference)−Encode axiomatic knowledge in rules and definitions

colleague(x,y)colleague(x,y) person(x) person(y) ((z group(z) z group(z) include(z,x) include(z,x) include(z,y)) include(z,y))

IFIF (rule for sufficient condition)(rule for sufficient condition) Group Include Person ?x?x Include Person ?y?y THENTHEN Person ?x?x ColleagueColleague Person ?y?y

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2020Summary of methodologySummary of methodology

Results−Design stages:

−In CoMMA this method provided O'CoMMA• 470 concepts (taxonomy depth = 13 subsumptions).

• 79 relations (taxonomy depth = 2 subsumptions).

• 715 terms in English and 699 in French.

• 550 definitions in English and 547 in French .

Data collection

Lexicons

RDFS version Algebraic properties Rules

Coverage ?

Structured tables

Scenarios

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2121Resulting ontology: O'CoMMAResulting ontology: O'CoMMA

Structure:

−Abstract top & middle layer for corporate memory: reusable−Middle layer for domain: reusable in same domain−Extension layer: usable but not reusable−Reuse tested e.g., CSELT CSTB, APROBATIOM, KMP

Top: Top: abstractabstract

Middle: Middle: common notionscommon notions

Extension: Extension: specificspecific

Organization Document Person Domain

dedicated to

corporate memory

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2222ProgressionProgression

PhD research: “D.A.I. and knowledge”Semantic intrawebs for corporate memories

Ontology engineering for semantic intrawebs

Multi-agent architecture for semantic intrawebs

Post-doc research: “D.A.I. and services”Semantic Web services &Context-awareness

Personal resource management & privacy enforcement

Experimental results

Wrap-up

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2323A multi-agent architecture for CoMMAA multi-agent architecture for CoMMA

Problem = handle information distribution−Handle naturally scattered information & knowledge −Assist diffusion of captured information and knowledge

Follow multi-agent paradigm:−Artificial societies collaborating for global capitalization−Artificial individual intelligence, able to locally adapt

Step 1 - Sub-societies identification−Started from the tasks to be performed

(provide ontology, manage annotations, manage users and matchmaking)

−Thus four sub-societies to handle these four tasks: Ontology & modelsub-society

User-dedicatedsub-society

Annotation-dedicatedsub-society

Matchmakerssub-society

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2424Organizing resource-dedicated sub-societiesOrganizing resource-dedicated sub-societies

Replicationresource-dedicated

Replication

Distribute resource onlyworkload & network load +/-redundancy +

Complete replicationredundant & less network load +content replicated –

Distribute roles & resourcedistribute workload +increase network load –

Cooperativeresource-dedicated

Peer-to-Peer

Representative

Resource-dedicated

Hierarchy

Step 2 - Analyse possible organization structures:

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2525Roles and interactionsRoles and interactions

Users' societyUsers' society

Annotations SocietyAnnotations SocietyOntology and Model SocietyOntology and Model Society

Interconnection SocietyInterconnection Society

Ontologist AgentsOntologist Agents

MediatorsMediators

ArchivistsArchivists

Profile Profile ManagersManagers

Profiles Profiles ArchivistsArchivists

InterfaceInterfaceControllersControllers

FederatedFederatedMatchmakersMatchmakers

Step 3 - Roles and interactions

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2626Zooming on annotation-dedicated sub-societyZooming on annotation-dedicated sub-society

Annotations SocietyAnnotations Society

MediatorsMediators

ArchivistsArchivists

The annotation-dedicated sub-society−Hierarchy:

Mediators & archivists

−Manage local sources of

annotations distributed

over the organization

network

Propose annotation-related services to others−Archive new annotations on documents of the memory−Search & retrieve references matching queries−Notify registered agents of the arrival of a new annotation

Present 2 roles and then 2 interaction protocols

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2727Annotation-dedicated sub-society (mediator)Annotation-dedicated sub-society (mediator)

role model Annotation Mediator role in the Annotation-dedicated society

responsibilities handle distribution of annotations over thearchivists both for new annotation submissionsand query solving processes

collaborators Directory facilitator, User Profile Manager,Ontology Archivist, Annotation Archivist,Corporate Model Archivist

external interfaces RDF annotation manipulation interfacerelationships -expertise query and submission managementinteractions Query-Ref, Contract-Net, Subscribe, Request;

FIPA ACLothers -

AM

directoryfacilitator

AA AA AA

Mediator role (supervises social interactions)

−Contact point for other sub-societies

−Supervising distribution of tasks for query solving

−Allocating a new annotation to an archive

−Notifying new annotation arrival to trigger push functions

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2828Annotation-dedicated sub-society (archivist)Annotation-dedicated sub-society (archivist)

role model Annotation Archivist role in the Annotation-dedicated society

responsibilities store and query the annotations of the memorycollaborators Directory facilitator, Annotation Mediator,

Ontology Archivistexternal interfaces RDF annotation manipulation interfacerelationships also part of the roles in Corporate Model

Archivist and User Profile Archivistexpertise annotation archiving an queryinginteractions Query-Ref, Contract-Net; FIPA ACLothers -

AA

annotationmediator

annotationarchive

Archivist role (manage a local archive)

−Attached to & exploits local base

−Answers to query as much as it can with local knowledge

−Proposes archiving services advertising the archive

content

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 2929Interaction mediator-archivist (allocation)Interaction mediator-archivist (allocation)

Interaction 1 - Annotation allocation− Pb: archives distributed all over organization− Mediator & archivists discuss best archive for new annot.− Contract-net (CfP, Proposal, Accept/Reject):

− Proposals: semantic distance new annotation - archive

1:cfp

2:cfp

2:cfp

3:propose

3:propose

:protocol fipa contract net:content <RDF Annotation>:language CoMMA-RDF:ontology CoMMA Ontology

5:accept/reject

:protocol fipa contract net:content <propose bid = distance archive / refuse / not understood>:language CoMMA-RDF:ontology CoMMA Ontology

4:propose

6:accept/reject

6:accept/reject

7:inform

7:inform8:informAMLocal:Med *:Med *:Arch

*:Arch

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3030Content of an annotation or of an archiveContent of an annotation or of an archive

Simple annotation <rdf:RDF xmlns:rdf="http:// (…) "> <CoMMA:ArticleArticle rdf:about="http://intranet/reports/R3029"> <CoMMA:TitleTitle>CfP UMTS AnalysisCfP UMTS Analysis</CoMMA:Title> <CoMMA:AuthorAuthor> <CoMMA:PersonPerson rdf:about="http://www.mycorp.com/~fab" /> </CoMMA:Author> </CoMMA:Article> </rdf:RDF>

Corresponding triples:

Unstructured set of triples to describe content type

Article

Author

Person

Article

Title

Literal: "CfP UMTS Analysis"

literal concept

Triple:

subject

predicate

object

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3131

Article

Title

Literal: "CfP UMTS Analysis"

Article

Author

Person

AnnotationAnnotation

ArchiveArchive

Distance between an annotation and an archiveDistance between an annotation and an archive

Ho

w c

lose

are

th

ey ?

Report

Title

Literal: "Negotiation in C-Net"

Literal: "Zeno paradox"......

31 31 instancesinstances

Report

Author

Person

27 27 instancesinstances

Report

Author

Committee

10 10 instancesinstances

Book

Author

Person

19 19 instancesinstances

Book

Title

Literal: "Agents for dummies"

Literal: "Franc and Euro"......

7 7 instancesinstances

StatisticsStatistics

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3232Building the distanceBuilding the distance

Case 1 - literal values: lexicographical distance Case 2 - concept types: minimum length of path

between two types through least common super type

DistH(Type1,Type2) = Min(GPath(Type1,LCST)+GPath(Type2,LCST)) GPath(,): number of edges through generalization links LCST: least common super type = shared characteristics

Distance between two triples (conditional weighted sum)

document

chart book

graphdiagram

LCST

Distance (diagram, graph) = 2

booklet

Distance (diagram, booklet)= 4

LCST

Dist TFStat(TripleA, TripleB) = DistC1+ DistR + DistC2

DistCi= WC* DistH(Type1,Type2) or WC* DistH(Type,Lit) or WL* N * DistI(Lit, [Blow,Bup]) with N=MaxC*2/MaxL

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3333Semantic distance between typesSemantic distance between types

Distance annotation archive

DistTStat(Triple,Stat) = Min(DistTFStat (Triple,Triplei)Triplei Stat

− triplestat:

Stat),(TripleDist )Stat ,An( Dist jTStatXAStat

Xj AnTriple

− annotationstat:

sub-type DistH = 0 Pref),(TripleDist )Pref ,An( Dist jTPrefXAPref

Xj AnTriple

− annotationpref:

Dist(AnX, ArcY) = DistAStat(AnX, StatY)+DistAPref(AnX, PrefY)− annotationarchive:

Distance = allocation criteria of contract-net− "and the winner is..." the archivist with smallest distance− Cluster annotations & specialize archives− Improve query solving & respect knowledge distribution

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3434Interactions mediator-archivist in solving a queryInteractions mediator-archivist in solving a query

Interaction 2 - distributed query solving− Pb: archives of annotations distributed all over org.− Mediators & archivists share knowledge to solve a query− Composition of Query-Ref protocol for distributed solving

1:query-ref

:protocol fipa query:content <RDF pattern / OBSIQ>:language CoMMA-RDF:ontology CoMMA Ontology

4:inform

2a:query-ref 3a:inform 2b:query-ref 3b:inform

5a,b,c,...:query-ref

5a,b,c,...:inform 5a,b,c,...:

query-ref

5a,b,c,...:inform

:protocol fipa query:content <RDF pattern /result>:language CoMMA-RDF:ontology CoMMA Ontology

LocalAM:Med *:Med

*:Arc*:Arc

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3535Overlap descriptionOverlap description

Each archivist calculates the overlap between query needs and its archive statistics:

− Archivists refine their service description for a query

− Mediators target multicast communications

− Exploit archive specialization obtained during the allocation of annotations

query needsquery needs archive statisticsarchive statistics&&

Description of the Description of the overlap need/archiveoverlap need/archive

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3636RDF query: tree structureRDF query: tree structure

<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:CoMMA="http://www.inria.fr/acacia/comma#"> <CoMMA:Document> <CoMMA:CreatedBy> <CoMMA:Person> <CoMMA:FamilyName>?AuthorName?AuthorName</CoMMA:FamilyName> <CoMMA:FirstName>?AuthorFirstName?AuthorFirstName</CoMMA:FirstName> </CoMMA:Person> </CoMMA:CreatedBy> <CoMMA:Title>?DocTitle?DocTitle</CoMMA:Title> <CoMMA:Edited> <CoMMA:Company> <CoMMA:Designation>~smith~smith</CoMMA:Designation> <CoMMA:Situated> <CoMMA:Location> <CoMMA:City>NiceNice</CoMMA:City> <CoMMA:Country>FranceFrance</CoMMA:Country> <CoMMA:Phone>?EditorPhone?EditorPhone</CoMMA:Phone> </CoMMA:Location> </CoMMA:Situated> <CoMMA:HasForActivity> <CoMMA:EditionEdition /> </CoMMA:HasForActivity> </CoMMA:Company> </CoMMA:Edited> <CoMMA:CreationDate>20002000</CoMMA:CreationDate> <CoMMA:ReviewedBy> <CoMMA:Person> <CoMMA:FamilyName>hofstadterhofstadter</CoMMA:FamilyName> <CoMMA:FirstName>douglasdouglas</CoMMA:FirstName> </CoMMA:Person> </CoMMA:ReviewedBy> <CoMMA:Concern> <CoMMA:ComputerScienceTopicComputerScienceTopic /> </CoMMA:Concern> </CoMMA:Document></rdf:RDF>

?

!

?

?

!!?

!

!

!

!

!

1

1

1

1

1

1

2

2

2

2

22

2

3

33

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3737DecompositionDecomposition

!!

!!!!

!!

!!

!!!!

!!

??

?? ??

??

Tree decomposition & work distribution− Mediator prepares query

(compact, cross-references) − Mediator decomposes query

• Bottom-up constraints solving

!!

!! !!

!!

!!

!! !!

!!

!!

!! !!

!!

!!

!! !!

!!

!!

!! !!

!!

!! !!!!

!!

!! !!

!!

!! !!!!

!!

!! !!

!! !!!!

!! !!

!! !!!! !! !!!!!! !! !!!!

!!

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3838DecompositionDecomposition

!!

!!!!

!!

!!

!!!!

!!

??

?? ??

??

Tree decomposition & work distribution− Mediator prepares query

(compact, cross-references) − Mediator decomposes query

• Bottom-up constraints solving• Top-down question answering• URI as cut/joint points

− Mediator sends subqueries to archivists according to overlap

??

????

??

??

????

??

!!!!

!!

??

????

??

!!!!

!!

??

!!

!!????

??

!!!!

!!

Forming final result− Mediator merges partial results− Mediator solves cross-references

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 3939Overview of agent-based designOverview of agent-based design

Design stages:

Result:• Society providing ontology and corporate model

• Annotation-dedicated society

• Two trials at mid-project and project end in development spiral

• One public open-day demo: industrial & EU commission

• Usability and usefulness recognized by users/public

Scenarios

Architectural requirements

Roles

Behaviours

Ontology

Interactions

Agent types

Configuration requirements

Deployment

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4040ProgressionProgression

PhD research: “D.A.I. and knowledge”Semantic intrawebs for corporate memories

Ontology engineering for semantic intrawebs

Multi-agent architecture for semantic intrawebs

Post-doc research: “D.A.I. and services”Semantic Web services &Context-awareness

Personal resource management & privacy enforcement

Experimental results

Wrap-up

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4141

(c) CMU - Sadehlab

Web service to humansWeb service to humans

To pay by check and/or register by mail, do not use this form! Please download and complete the PDF form instead. You can also complete the PDF version of the form electronically, and then print it and either fax or mail the printout to AAAI. (Note: You can only save your data entry if you have Acrobat 4.0 or higher. Acrobat Reader allows you to complete the form and print it, but not save it.) Student registrants please note that you are still required to send hard copy proof of your full-time student status to the AAAI office in order to receive the discounted student registration rates.

First Name (and optional middle initial):Last Name (family or surname):Company or Affiliation:Mailing Address:City: State:USA Zip Code (Please enter your zip+4 code):Country: Foreign Postal Code:Work or Daytime Telephone:E-mail Address:Please select your tutorial(s) / workshop(s) and select the rateSunday AM Sunday PM Monday AM Monday PM $100* $300*To qualify for this price, you must also register for the IJCAI technical conference above.

Credit card number:

submit reset

T8 T8 W2 W2

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4242

(c) CMU - Sadehlab

Web service to machinesWeb service to machines

Fie eio uaz edaziajqz b azeiqns aeo qzjioqs ef loedfze euf azekq,e f;qzejfbqzeuqei”çzerbé’rèé” b a”rçé”rb & “rç”elfrmqz$eé”r aoenbqzrelaze’r^)é’ramz’ pùze aé”)r)é”rnze,r zeqùf$ze qze, rtaoez qzr;,zep^é”rnq :zepqzerlg q;ze, apzé”’,’r zerppé”ar na “rier “ é”_efnf f&_”’kn fnàe-o”’n’rà”’n n fàozrnzero”nkn’zené” rz ràzerié l”, é”iozlt;,zer lserpqmrezneiohljké’é”’( nt tgaé”inn ze’rtnoser tao’tzket nzerot ite av ioit tngoierg aert_niaoit naeriaz azg pzaena zetpaergn rega’tioaernze rtszergiooiz’tnlqlksrngznqsregpiozernge nrgodsf/

Kz zi’èé 56 b’_é é”ç’ zeiN”u’z9 “ufkzedzr82 “iz aé”iusz^fkeuéa§ ué d8 ér8 zejé= é1ù^geddez’pzrE9mMa£7nSù Krµù1r 5yr kr9jz zIO £ 17v UIZ1 bzz di1µ*§jdea »zzehze2¨r0pa1l¨/$r ¨&7bziY+=ç°+=|r8_#zf&ze!Mq8^ae%rà/§µaz.lr6l%µ£../§!6o=rI « z£9 zj?ai1çd »eazo LazzOçù§de*!aqµdaeo zoaz azozi8£&à9kzeai %§ 9kzeai Z¨ pµa£m %§ pµa£m Z¨ {ass] {µss]%deK;?9 dzeoPSceze £1zé %Azµdza^0çNéçqefnqEnioUZiunc IOzazOxdjsu.

é#%µeio%MzefzIAr

“£§ya è0°&

è° é° *ù *ù

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4343

(c) CMU - Sadehlab

Positioning (semantic Web services)Positioning (semantic Web services)

Internet & Web (shared service-sphere for humans)

Semantic Web (shared infosphere for machines)

Semantic Web Services (shared service-sphere for machines)

SOAP, XMLP, WSDL, DAMLS, UDDI (v2 v3)

Hyp: Semantic web services for e-Businessand Enterprise Applications Integration

Hyp: OntologyOntology to support semantic Web services

Distributed artificial intelligence−Intelligent Agent Intelligent Services−Multi-agent systems Service interaction & composition

Hyp: Multi-agent architectures for webs of services

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4444

(c) CMU - Sadehlab

Interface without context awarenessInterface without context awareness

To pay by check and/or register by mail, do not use this form! Please download and complete the PDF form instead. You can also complete the PDF version of the form electronically, and then print it and either fax or mail the printout to AAAI. (Note: You can only save your data entry if you have Acrobat 4.0 or higher. Acrobat Reader allows you to complete the form and print it, but not save it.) Student registrants please note that you are still required1) to send hard copy proof of your full-time student status to the AAAI office in order to receive the discounted student registration rates.

First Name (and optional middle initial):Last Name (family or surname):Company or Affiliation:Mailing Address:City: State:USA Zip Code (Please enter your zip+4 code):Country: Foreign Postal Code:Work or Daytime Telephone:E-mail Address:Please select your tutorial(s) / workshop(s) and select the rateSunday AM Sunday PM Monday AM Monday PM $100* $300*To qualify for this price, you must also register for the IJCAI technical conference above.

Credit card number:

submit reset

T8 T8 W2 W2

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4545

(c) CMU - Sadehlab

Interface with context awarenessInterface with context awareness

Please select your tutorial(s) / workshop(s) and ($100 rate)Sunday AM Sunday PM Monday AM Monday PM

Just do it !

T8 T8 W2 W2

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4646

(c) CMU - Sadehlab

Positioning (mobile access to SW&S)Positioning (mobile access to SW&S)

Mobile (net)working−Mobile networks:

• Phones / pagers PDAs, sub-notebooks

• Telecom services Web and online services

−Mobile accesses reduced & constrained interactions (interfaces, connectivity, cognitive workload)

−Schism: service availability & user availability

Hyp: Context-awareness to support mobile access to Context-awareness to support mobile access to semantic Webs and their servicessemantic Webs and their services

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4747

(c) CMU - Sadehlab

Positioning (context awareness)Positioning (context awareness)

Context awareness State of the Art−Application leveraging context awareness

Active Badge [Want et al., 92] ParcTab [Schilit, 95] Oxygen

[Dertouzos, 99] GUIR [Hong & Llanday, 01] Aura [Garlan et al., 02] • Application dependent and heterogeneous

• Redundant and scattered

−Personal resources integration and unification• Toolkits and widgets for wrapping [Dey et al., 00]

• e-Wallet: awareness & privacy You are here

Hyp: Semantic Web & Services to provide a unified Semantic Web & Services to provide a unified secure interface to personal resources (e-Wallet)secure interface to personal resources (e-Wallet)

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4848

(c) CMU - Sadehlab

Context of researchContext of researchmyCampus: a context-aware environment aimed at

enhancing access to services for everyday campus life at Carnegie Mellon University (CMU)

− BBN, IBM, HP, Symbol and Boeing − Air Force Research Laboratory (contract F30602-02-2-0035)

− Defense Advanced Research Project Agency (DARPA) (contract F30602-98-2-0135)

Interactions with− SONAT: user-aware notification (D.o.D.)

− CoSAR (I-X, KAoS/CoABS Grid): notification planning (AIAI)

− SWAP: Semantic Web and Peer-to-peer for KM (IST program)

My focus: Semantic Web & Services, Context-awareness and Privacy, M.A.S. and Ontologies

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 4949

(c) CMU - Sadehlab

Open architecture – mobile access Open architecture – mobile access

PDA & Wireless NetworkAgent roles:

− Platform manager− User interaction

manager− Growing collection of

task-specific agents− e-Wallet manager

Web resources− Semantic Web services− Semantic Web ontologies− Semantic Web annotations− Search engines

SemanticSemanticWeb serversWeb servers

S. Web Ontologies

S. Web AnnotationsOther Web Resources

Semantic Web Services

SemanticSemanticsearchsearchservicesservices

Task-Task-specificspecific

Task-specific resources and APIs

Communication toolkit (http, e-mail, IM, etc.)

User interaction User interaction managermanager

API

NETWORK

PlatformPlatformmanagermanager

White & yellow pages

MAS administration toolkit

knowledge base

API

e-Wallet e-Wallet managermanager

Security toolkitAPI

Web services invocation toolkitAPI

Inference engineAPIe-Wallete-Wallet

knowledge base

Service activation rulesDynamic knowledge about owner

Static knowledge about owner

Loaded ontologies

Privacy enforcement rules

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5050

(c) CMU - Sadehlab

ProgressionProgression

PhD research: “D.A.I. and knowledge”Semantic intrawebs for corporate memories

Ontology engineering for semantic intrawebs

Multi-agent architecture for semantic intrawebs

Post-doc research: “D.A.I. and services”Semantic Web services &Context-awareness

Personal resource management & privacy enforcement

Experimental results

Wrap-up

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5151

(c) CMU - Sadehlab

e-Wallet & the knowledge typologye-Wallet & the knowledge typology

Secured unified semantic interface to access knowledge about the e-Wallet owner

Static assertional knowledge:− User’s static profile (name, SSN, position, etc.)− Static contextual knowledge (campus description, etc.)

fetched from outside as needed.

Dynamic assertional knowledge:− User’s dynamic profile (contextual preferences, etc.)− Context knowledge (location, current activity, weather,

etc.) fetched from outside as needed.− Privacy: authorization & obfuscation (precision, lie, etc.)

Ontological knowledge, fetched from an outside repository of ontology at startup.

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5252

(c) CMU - Sadehlab

privacy

query

answer

e-e-

Design of an e-WalletDesign of an e-Wallet

Three-layer architecture− Core knowledge: static & dynamic knowledge of user− Service Layer: invoke external sources

of knowledge: web services andpersonal resources

− Privacy layer: enforce privacy ruleson external requests: access,obfuscation (precision, lies)

Asserting elementary needs for authorized information

Pre-check access rights

Post-checkaccess rights

Fetch usefulstatic knowledge

Application ofobfuscation rules

Query contextassertionQuery

Assertion ofauthorized knowledge

Result

Call relevant external services

service

CoreKnow-ledge

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5353

(c) CMU - Sadehlab

Processes in the e-WalletProcesses in the e-Wallet

OWLMeta-model

in CLIPS

OWLMeta-model

in CLIPS

Ontologyin OWL

Ontologyin OWL

Annotationin OWL

Annotationin OWL

Rulein (R)OWL

Rulein (R)OWL

Servicesin (W)OWLServices

in (W)OWL

Securityin (S)OWLSecurity

in (S)OWL

Queryin (Q)OWL

Queryin (Q)OWL

OntologystylesheetOntologystylesheet

&

AnnotationstylesheetAnnotationstylesheet

&

Rulestylesheet

Rulestylesheet

&

Servicestylesheet

Servicestylesheet

&

SecuritystylesheetSecurity

stylesheet&

Querystylesheet

Querystylesheet

&

Ontologyin CLIPSOntologyin CLIPS

Annotationin CLIPS

Annotationin CLIPS

Rulein CLIPS

Rulein CLIPS

Service rulein CLIPS

Service rulein CLIPS

Security rulein CLIPS

Security rulein CLIPS

Query rulesin CLIPS

Query rulesin CLIPS

XSLT Engine

Resultin OWLResultin OWL

JESS

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5454

(c) CMU - Sadehlab

e-Wallet core: JESS inference enginee-Wallet core: JESS inference engine

CLIPS language:− Atoms, numbers, strings, functions, variables,

Java reflection− Ordered facts (not used here)− Unordered facts (with templates // classes)

− Forward rules

− Queries: special kind of rule with no right-hand-side

− Declare backward chaining reactive rules using(do-backward-chaining …) function(template need-template)

− The RETE algorithm

(person (name "Adeline") (age 20) (gender Female))(person (name "Adeline") (age 20) (gender Female))

(defrule adult ?p <- (person (age >=18)) => (assert (adult ?p))(defrule adult ?p <- (person (age >=18)) => (assert (adult ?p))

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5555

(c) CMU - Sadehlab

e-Wallet semantic enginee-Wallet semantic engine RDF Triple model

RDFS & OWL meta-model (e.g, symmetry of properties)

(deftemplate triple "Template representing a RDF triple" (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

(deftemplate triple "Template representing a RDF triple" (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

Triple: (predicate, subject, object)Triple: (predicate, subject, object)

<rdfs:Class rdf:ID="SymmetricProperty"> <rdfs:label>SymmetricProperty</rdfs:label> <rdfs:subClassOf rdf:resource="#ObjectProperty"/></rdfs:Class>

<rdfs:Class rdf:ID="SymmetricProperty"> <rdfs:label>SymmetricProperty</rdfs:label> <rdfs:subClassOf rdf:resource="#ObjectProperty"/></rdfs:Class>

(triple (predicate "http://www.w3.org/2000/01/rdf-schema#subClassOf") (subject "http://www.w3.org/2002/07/owl#SymmetricProperty") (object "http://www.w3.org/2002/07/owl#ObjectProperty") )(defrule symmetry (declare (salience 100)) (triple (predicate "http://www.w3.org/1999/02/22-rdf-syntax-ns#type") (subject ?p) (object "http://www.w3.org/2002/07/owl#SymmetricProperty")) (triple (predicate ?p) (subject ?x) (object ?y))=> (assert (triple (predicate ?p) (subject ?y) (object ?x))) )

(triple (predicate "http://www.w3.org/2000/01/rdf-schema#subClassOf") (subject "http://www.w3.org/2002/07/owl#SymmetricProperty") (object "http://www.w3.org/2002/07/owl#ObjectProperty") )(defrule symmetry (declare (salience 100)) (triple (predicate "http://www.w3.org/1999/02/22-rdf-syntax-ns#type") (subject ?p) (object "http://www.w3.org/2002/07/owl#SymmetricProperty")) (triple (predicate ?p) (subject ?x) (object ?y))=> (assert (triple (predicate ?p) (subject ?y) (object ?x))) )

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5656

(c) CMU - Sadehlab

e-Wallet semantic enginee-Wallet semantic engine

Ontologies: (e.g., declare man, location, etc.)

Annotations: (e.g., Fabien is in Smith Hall, etc.)

<owl:Class rdf:ID="Person"> <rdfs:subClassOf rdf:resource="#Entity" /></owl:Class><owl:ObjectProperty rdf:ID="location"> <rdfs:domain rdf:resource="#Entity" /> <rdfs:range rdf:resource="#Place"/></owl:ObjectProperty>

<owl:Class rdf:ID="Person"> <rdfs:subClassOf rdf:resource="#Entity" /></owl:Class><owl:ObjectProperty rdf:ID="location"> <rdfs:domain rdf:resource="#Entity" /> <rdfs:range rdf:resource="#Place"/></owl:ObjectProperty>

(triple (predicate "http://www.w3.org/2000/01/rdf-schema#subClassOf") (subject "http://sadehlab.cs.cmu.edu/mycampus#Person") (object "http://sadehlab.cs.cmu.edu/mycampus#Entity")) ...

(triple (predicate "http://www.w3.org/2000/01/rdf-schema#subClassOf") (subject "http://sadehlab.cs.cmu.edu/mycampus#Person") (object "http://sadehlab.cs.cmu.edu/mycampus#Entity")) ...

<mc:Man rdf:ID="http://cs.cmu.edu/People/~fgandon"> <mc:location rdf:resource="http://cmu.edu/SmithHall"/></mc:Man>

<mc:Man rdf:ID="http://cs.cmu.edu/People/~fgandon"> <mc:location rdf:resource="http://cmu.edu/SmithHall"/></mc:Man>

(triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#location") (subject "http://cs.cmu.edu/People/~fgandon") (object "http://cmu.edu/SmithHall")) ...

(triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#location") (subject "http://cs.cmu.edu/People/~fgandon") (object "http://cmu.edu/SmithHall")) ...

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5757

(c) CMU - Sadehlab

e-Wallet semantic enginee-Wallet semantic engine Rules: (e.g., when in I am in a meeting I am busy)

<rowl:Rule direction="forward"> <rdfs:label>Meeting means busy</rdfs:label> <rowl:head> <mc:Person rdf:ID="&variable;#person"> <mc:availability><mc:Busy rdf:ID="&mc;#Busy"/></mc:availability> </mc:Person> </rowl:head> <rowl:body> <mc:Person rdf:ID="&variable;#person"> <mc:activity> <mc:Meeting rdf:ID="&variable;#activity"/> </mc:activity> </mc:Person> </rowl:body> </rowl:Rule>

<rowl:Rule direction="forward"> <rdfs:label>Meeting means busy</rdfs:label> <rowl:head> <mc:Person rdf:ID="&variable;#person"> <mc:availability><mc:Busy rdf:ID="&mc;#Busy"/></mc:availability> </mc:Person> </rowl:head> <rowl:body> <mc:Person rdf:ID="&variable;#person"> <mc:activity> <mc:Meeting rdf:ID="&variable;#activity"/> </mc:activity> </mc:Person> </rowl:body> </rowl:Rule>

(defrule Meeting-means-busy ... (triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#activity") (subject ?person) (object ?activity)) (triple (predicate "http://www.w3.org/1999/02/22-rdf-syntax-ns#type") (subject ?activity) (object "http://sadehlab.cs.cmu.edu/mycampus#Meeting") ) => ...(assert (triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#availability") (subject ?person)(object "http://sadehlab.cs.cmu.edu/mycampus#Busy")

(defrule Meeting-means-busy ... (triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#activity") (subject ?person) (object ?activity)) (triple (predicate "http://www.w3.org/1999/02/22-rdf-syntax-ns#type") (subject ?activity) (object "http://sadehlab.cs.cmu.edu/mycampus#Meeting") ) => ...(assert (triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#availability") (subject ?person)(object "http://sadehlab.cs.cmu.edu/mycampus#Busy")

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5858

(c) CMU - Sadehlab

Service and privacy layers− Special types of triples− Backward chaining reaction started by the query

• Privacy rules (clearance/revision)• Service rules (dynamic access)• Static migration rules

e-Wallet semantic enginee-Wallet semantic engine

inference engine

privacy

service

coreknow-ledge

query

answer

(deftemplate authorized_triple (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

(deftemplate authorized_triple (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

(deftemplate service_triple (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

(deftemplate service_triple (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

(deftemplate triple (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

(deftemplate triple (slot predicate (default "")) (slot subject (default "")) (slot object (default "")))

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 5959

(c) CMU - Sadehlab

QueryQuery

<qowl:Query rdf:ID=""> <qowl:sender rdf:resource="http://cs.cmu.edu/~nsadeh"/></qowl:Query><mc:Person rdf:ID="http://cs.cmu.edu/~fgandon"> <mc:location rdf:resource="&variable;#location" /></mc:Person>

<qowl:Query rdf:ID=""> <qowl:sender rdf:resource="http://cs.cmu.edu/~nsadeh"/></qowl:Query><mc:Person rdf:ID="http://cs.cmu.edu/~fgandon"> <mc:location rdf:resource="&variable;#location" /></mc:Person>

(defrule query (declare (salience 0)) … (authorized_triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#location") (subject "http://cs.cmu.edu/~fgandon") (object ?location)) => (store-result location ?location))

(defrule query (declare (salience 0)) … (authorized_triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#location") (subject "http://cs.cmu.edu/~fgandon") (object ?location)) => (store-result location ?location))

Query issued by the user 'nsadeh’ requesting the location of user ‘fgandon’Query issued by the user 'nsadeh’ requesting the location of user ‘fgandon’

Three steps:− Query context assertion− Query rule definition

• Body: request for authorized triples• Head: storage & pretty printing function

− Rule execution

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6060

(c) CMU - Sadehlab

Privacy rulesPrivacy rules<sowl:ReadAccessRule> <rdfs:label>people can only know whether or not I am on campus</rdfs:label> <sowl:target> <mc:Person rdf:ID="&variable;#owner"> <mc:location rdf:resource="&variable;#location"/> </mc:Person> </sowl:target> <sowl:check> <rowl:And> <rowl:condition> <mc:E-Wallet rdf:ID="&variable;#e-Wallet"> <mc:owner> <mc:Person rdf:ID="&variable;#owner"/> </mc:owner> </mc:E-Wallet> </rowl:condition> <rowl:condition> <mc:Place rdf:ID="http://www.cmu.edu"> <mc:include rdf:resource="&variable;#location" /> </mc:Place> </rowl:condition> <rowl:not-condition> <qowl:Query rdf:ID="&variable;#query"> <qowl:sender rdf:resource="&variable;#owner" /> </qowl:Query> </rowl:not-condition> </rowl:And> </sowl:check> <sowl:revision> <mc:Person rdf:ID="&variable;#owner"> <mc:location rdf:resource="http://www.cmu.edu"/> </mc:Person> </sowl:revision></sowl:ReadAccessRule>

<sowl:ReadAccessRule> <rdfs:label>people can only know whether or not I am on campus</rdfs:label> <sowl:target> <mc:Person rdf:ID="&variable;#owner"> <mc:location rdf:resource="&variable;#location"/> </mc:Person> </sowl:target> <sowl:check> <rowl:And> <rowl:condition> <mc:E-Wallet rdf:ID="&variable;#e-Wallet"> <mc:owner> <mc:Person rdf:ID="&variable;#owner"/> </mc:owner> </mc:E-Wallet> </rowl:condition> <rowl:condition> <mc:Place rdf:ID="http://www.cmu.edu"> <mc:include rdf:resource="&variable;#location" /> </mc:Place> </rowl:condition> <rowl:not-condition> <qowl:Query rdf:ID="&variable;#query"> <qowl:sender rdf:resource="&variable;#owner" /> </qowl:Query> </rowl:not-condition> </rowl:And> </sowl:check> <sowl:revision> <mc:Person rdf:ID="&variable;#owner"> <mc:location rdf:resource="http://www.cmu.edu"/> </mc:Person> </sowl:revision></sowl:ReadAccessRule>

Privacy rule: grant access to location when on campus but obfuscate precisionPrivacy rule: grant access to location when on campus but obfuscate precision

http://www.at.home.gandon.fr/

truth

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6161

(c) CMU - Sadehlab

Service rulesService rules<wowl:ServiceRule wowl:salience="50"> <rdfs:label>provide location for IP Address</rdfs:label> <wowl:output> <mc:Entity rdf:ID="&variable;#entity"> <mc:location rdf:resource="&variable;#location" /> </mc:Entity> </wowl:output> <wowl:precondition> <mc:Entity rdf:ID="&variable;#entity"><mc:ip>&variable;#ip</mc:ip> </mc:Entity> </wowl:precondition> <wowl:call> <wowl:Service wowl:name="call-web-service"> <wowl:parameter>http://cmu.edu/location_tracking#</wowl:parameter> <wowl:parameter>&variable;#ip</wowl:parameter> </wowl:Service> </wowl:call></wowl:ServiceRule>

<wowl:ServiceRule wowl:salience="50"> <rdfs:label>provide location for IP Address</rdfs:label> <wowl:output> <mc:Entity rdf:ID="&variable;#entity"> <mc:location rdf:resource="&variable;#location" /> </mc:Entity> </wowl:output> <wowl:precondition> <mc:Entity rdf:ID="&variable;#entity"><mc:ip>&variable;#ip</mc:ip> </mc:Entity> </wowl:precondition> <wowl:call> <wowl:Service wowl:name="call-web-service"> <wowl:parameter>http://cmu.edu/location_tracking#</wowl:parameter> <wowl:parameter>&variable;#ip</wowl:parameter> </wowl:Service> </wowl:call></wowl:ServiceRule>

(defrule provide-location-for-IP-Address (declare (salience 50)) … (need-dynamic_triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#location") (subject ?entity) (object ?location) ) … => (call-web-service "http://cmu.edu/location_tracking#" ?ip))

(defrule provide-location-for-IP-Address (declare (salience 50)) … (need-dynamic_triple (predicate "http://sadehlab.cs.cmu.edu/mycampus#location") (subject ?entity) (object ?location) ) … => (call-web-service "http://cmu.edu/location_tracking#" ?ip))

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6262

(c) CMU - Sadehlab

ProgressionProgression

PhD research: “D.A.I. and knowledge”Semantic intrawebs for corporate memories

Ontology engineering for semantic intrawebs

Multi-agent architecture for semantic intrawebs

Post-doc research: “D.A.I. and services”Semantic Web services & Context-awareness

Personal resource management & privacy enforcement

Experimental results

Wrap-up

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6363

(c) CMU - Sadehlab

Prototype #1: proof of conceptPrototype #1: proof of concept

Mockup of multi-agent architecture Two task-specific agents

− Restaurant Concierge Agent RCA(location, weather, food, price, time)

− Message Filtering Agent(interests, current activity) + Jabber (Instant Messaging)

Limited version of e-Wallet JSP pages to simulate User Interface Manager Three Web services

− Location tracking− Calendar access− Weather report

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6464

(c) CMU - Sadehlab

logginglogging

preferencespreferences

Setting preferencesSetting preferences

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6565

(c) CMU - Sadehlab

Using servicesUsing services

RestaurantRestaurantConciergeConcierge

MessageMessagefeedbackfeedback

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6666

(c) CMU - Sadehlab

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6767

(c) CMU - Sadehlab

Experiment #1 with early prototypeExperiment #1 with early prototype

Before the experiment:− Office for Human Research Protections approval

Institutional Review Board certificate− Selected group of 11 users, with a variety of profiles− Trained the 11 users during a 2-hour session + material

The 3-day experiment involved: − Message filtering agent: 44 messages and 484 feedbacks− Restaurant concierge agent: 28 recommendations− Logs were generated for each one of these events.

After the experiment:− Users had to fill a survey ~1/2 hour− Face-to-face de-briefing interviews ~15 minutes− Statistics on the logs

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6868

(c) CMU - Sadehlab

Experiment #1 extracts of feedback analysisExperiment #1 extracts of feedback analysis

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

(1) Bomb (2) Box Office (3) Songwriters (4) Consulting (5) Forecast (6) Showtimes (7) Symphony (8) Taxes

Any time

Available

Instantly

No interest

Undecided

0,00%

10,00%

20,00%

30,00%

40,00%

50,00%

60,00%

70,00%

80,00%

90,00%

100,00%

(1) Bomb (2) Songw riters (3) Forecast (4) Jamaica (5) Women - Research

Male

Female

0,00%

10,00%

20,00%

30,00%

40,00%

50,00%

60,00%

70,00%

80,00%

90,00%

100,00%

(1) technology conference (2) bomb (3) Cell-Matrix Adhesion (4) Skin of teeth (5) Psychology Relationship

Science

Humanities

Technology

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 6969

(c) CMU - Sadehlab

Experiment #1 most important resultsExperiment #1 most important resultsFor the Restaurant Concierge, 12.5 %

recommendations accepted thanks tocontext-awareness = 14.29% improvement.

For the Messaging: filtering and routing based on static profile = too loose.

− finer knowledge about theusers and finer filtering criteria;

− ~70% of the messages benefitfrom context-aware filteringand routinge.g. messages to be sent onlywhen available or when the day is over

Need critical mass of content, users and useful services

OK 61,36%

Too loose

32,23%

Too strict

6,40%

OK 61,36%

Too loose

32,23%

Too strict

6,40%

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 7070

(c) CMU - Sadehlab

Prototype #2: study of a useful servicePrototype #2: study of a useful service Focus:

− Find out about interesting events & Get information to people who want it

− Existing systems to distribute information? Design Computer Science Information Sc. Masters Electrical

Eng.

POSTERSPOSTERS

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 7171

(c) CMU - Sadehlab

Experiment #2 H.C.I. study of a useful serviceExperiment #2 H.C.I. study of a useful service

Improving messaging for events:Improving messaging for events:− Tag each poster with a list of

locations (situated)

− Display poster when user isclose to it or explicitly interested in the topic (distribute)

− Peer-to-peer publishing(person-2-person)

− Integrate with calendar(remember)

− Stop publishing when outdated (maintenance)

− Under evaluation; good feedback

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 7272ProgressionProgression

PhD research: “D.A.I. and knowledge”Semantic intrawebs for corporate memories

Ontology engineering for semantic intrawebs

Multi-agent architecture for semantic intrawebs

Post-doc research: “D.A.I. and services”Semantic Web services &Context-awareness

Personal resource management & privacy enforcement

Experimental results

Wrap-up

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 7373Wrap-up: conclusionWrap-up: conclusion

Ph.D. contributions− Methodologies and tools to build ontology, semantic

intraweb and associated multi-agent architecture

− A running prototype as a proof of concept for each point

− Both are reusable contributions

− Current extensions: Web scrappers & Semantic gateways

Post-Doc contributions ( as a continuation)

− Design M.A.S. architecture for Semantic Web Services with a focus on mobile accesses

− Use of S.W. and ontologies to allow context-awareness

− Integration S.W. and ontologies in privacy enforcement

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 7474Wrap-up: discussionWrap-up: discussion

Research ahead (remain at the state of the art)− Ontology and Memory life cycles (collective-ware, coherence)

− Semantic Webs (semantic mapping, open/extra, community, security)

− Semantic Services and Agents (EAI, e-Sourcing, Grid, autonomic)

− Mobile & ubiquitous Semantic Web (connectivity, pervasiveness)

− Intelligent user interfaces (customization, awareness, ergonomics)

Perfectly integrated with research of ACACIA & INRIA

D.A.I. & Web ServicesD.A.I. & Web Services

Ontology &Ontology &Knowledge representationKnowledge representation

Semantic WebSemantic Web& Knowledge management& Knowledge management

formalisms &applications

methods &schemata

formalisms,semantic ground forinferences & communication

software platforms for life-cycle

formalisms,annotated worlds,

methods, cases

software architectures,deployment platforms

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 7575AcknowledgmentsAcknowledgments

ACACIA Laboratory - INRIA Sophia Antipolis− Dr. Rose Dieng-Kuntz (Research Director – ACACIA project leader)

− Members of ACACIA team and logistics of INRIA Sophia− ATOS-Origin, CSTB, Deutsch Telekom T-Nova, Italia

Telecom, LIRMM and University of Parma − IST Program (CoMMA project)

School of computer science – Carnegie Mellon Uni.− Prof. Norman M. Sadeh (Mobile Commerce Laboratory Director,

SCS CMU, Free University of Amsterdam, European Commission)

− Members of myCampus team and logistic of ISRI− BBN, IBM, HP, Symbol and Boeing − IST Program (SWAP project)− Air Force Research Laboratory (contract F30602-02-2-0035)

− Defense Advanced Research Project Agency (DARPA) (contract F30602-98-2-0135)

(c) CMU - Sadehlab

Demo: CoMMA Ontology interfaceDemo: CoMMA Ontology interface

(c) CMU - Sadehlab

CoMMA: profile-based retrievalCoMMA: profile-based retrieval

Demo: interface CoMMA Demo: interface CoMMA

specified by INRIA, implemented by ATOS-Originspecified by INRIA, implemented by ATOS-Origin

Demo: CoMMA annot. allocation interactionsDemo: CoMMA annot. allocation interactions

cfp

bids

result

1 mediator 3 archivists

Demo: CoMMA query solving interactionsDemo: CoMMA query solving interactions

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 8181Demo: InfoBridge virtual postersDemo: InfoBridge virtual posters

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 8282

Fabien.Gandon@cs.cmu.edu / Fabien.Gandon@cs.cmu.edu / 8383AcknowledgmentsAcknowledgments

ACACIA Laboratory - INRIA Sophia Antipolis− Dr. Rose Dieng-Kuntz (Research Director – ACACIA project leader)

− Members of ACACIA team and logistics of INRIA Sophia− ATOS-Origin, CSTB, Deutsch Telekom T-Nova, Italia

Telecom, LIRMM and University of Parma − IST Program (CoMMA project)

School of computer science – Carnegie Mellon Uni.− Prof. Norman M. Sadeh (Mobile Commerce Laboratory Director,

SCS CMU, Free University of Amsterdam, European Commission)

− Members of myCampus team and logistic of ISRI− BBN, IBM, HP, Symbol and Boeing − IST Program (SWAP project)− Air Force Research Laboratory (contract F30602-02-2-0035)

− Defense Advanced Research Project Agency (DARPA) (contract F30602-98-2-0135)