The semantics of dialogue acts. Harry Bunt Oxford, IWCS 2011. The semantics of dialogue acts bridging the gap between computational semantics and pragmatics. Harry Bunt Oxford, IWCS 2011. Dialogue acts: Basics
An Image/Link below is provided (as is) to download presentationDownload Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.Content is provided to you AS IS for your information and personal use only. Download presentation by click this link.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.During download, if you can't get a presentation, the file might be deleted by the publisher.
E N D
Presentation Transcript
The semantics of dialogue acts
Harry Bunt Oxford, IWCS 2011
The semantics of dialogue actsbridging the gap betweencomputationalsemantics and pragmatics
Harry Bunt Oxford, IWCS 2011
Dialogue acts: Basics DIT, DIT++, and ISO 24617-2 DiAML: Dialogue Act Markup Language Context modelling Context model update semantics for dialogue acts Conclusions and perspectives Plan
Why don’t you start? question/suggestion Do you know what time it is? question/reproach It’s nearly five o’clock. statement/answer/warning/reproach Action-based semantics: dialogue acts asaddressees’ information-state updateoperators. (or ‘context update’) Dialogue acts have a communicative function and a semanticcontent (propositional/referential/action-al); the communicative function specifies a type of update operation; the semantic content is the material for updating the IS with. A communicative function can be applied to a semantic content in order to produce a dialogue act. Study of the semantics of communicative functions complements traditional compositional semantics Dialogue acts
Speech Act Theory Communication as Cooperation (Grice) (Austin, Searle) Communicative Activity Analysis (Allwood) ( HCRC TRAINS MRDA … GBG-IM DIT Verbmobil DAMSL MATE DIT++ LIRICS ISO 24617-2 (DIS) DIT++ Release 5 Dialogue act analysis: current state
Participating in a dialogue involves more than ‘just’ pursuing a certain goal, task or activity: Giving and eliciting feedback Taking turns Managing the use of time Establishing and maintaining contact Dealing with social obligations (greeting, thanking, apologizing,…) ….. Communication has many‘dimensions’ Anutterancemay have partsoraspectsthat have a communicativefunctionrelating to different dimensions – bemultifunctional Multidimensionality
Dimensions for dialogue acts DIT Definitionof a multimensionaltaxonomy of dialogue acts: Set of dialogue acts, partitionedinto clusters of whicheachmemberaddresses the same aspect of communication,and whereeach of these aspectscanbeaddressedindependently of the otherdimensions. (In otherwords, dimensions are ‘independent’ or ‘orthogonal’.) DIT: Everytwocommunicativefunctionswhichcanbeused in a certaindimension are eithermutuallyexclusiveorone is a specialization of the other (e.g.AnswerConfirmation) a functional segment can have a communicativefunction in each dimension,butnever more thanone (moduloentailedfunctions)
DIT: 10 dimensions of communication Task: Performing a certain ‘task’ or ‘activity’ through the dialogue Feedback: auto-feedback; providinginformationaboutone’sprocessing (perception, understanding, evaluation,…) of previousutterances; allo-feedback: assertingorelicitinginformation the partner’s processing of previousutterances Interaction Management: managing contact turnallocation(speaker role) use of time structuringof the discourse editingof one’sownand ofpartner’s speech socialobligations- greeting, thanking, apologising, sayinggoodbye,… All these types of information must berepresented in informationstates for ISU semantics of dialogue acts !
DIT++ taxonomy of communicative functions Communicativefunctions: specificforparticulardimensions, e.g. Turn Grab, Stalling, Apology,...: “dimension-specificfunctions” “general-purposefunctions”, applicable to anydimension of communication, e.g.YN-Question, Inform, Request,...: Total of 93 (basic) communicativefunctions: - 33 general-purposefunctions - 60 dimension-specificfunctions + ‘functionqualifiers’ forvariationdue to uncertainty, conditionality, sentiment See http://dit.uvt.nl
Communicative function qualifiers Variationsw.r.t. uncertainty, conditionality, incompleteness, and sentiment, e.g.: A: Wouldyoulike to have somecoffee? B1: Onlyifyou have itready. (Conditional Accept Offer) B2: Thatwouldbewonderful! (Happily Accept Offer) A: Do youknowwhat time the meeting will end? B: Notsure, maybesomethingliketwoo’clock. (UncertainAnwer)
ISO standard 24617-2 (DIS) Dialogue act annotation Slightlysimplifiedversion of DIT++ taxonomy Communicativefunctionsdefined as ISO data categories, following ISO standard12620 (seehttp://semantic-annotation.uvt.nl/) AnnotationlanguageDiAML(Dialogue Act MarkupLanguage), with abstract syntax + formalsemantics and concrete XML-basedsyntax Annotatedmultilingual test suites (English, Dutch, Italian) Established as Draft International Standard, January11, 2011 Expected to become ISO standard later in 2011
ISO Project 24617-2 Project team: Jan Alexandersson Harry Bunt (PL) Jean Carletta Alex ChengyuFangJae-WoongChoeKoitiHasida VolhaPetukhovaAndreiPopescu-Belis Claudia Soria David Traum Consultinggroup: Jens Allwood James Allen Nick Campbell RobertaCatizone Anna EspositoThierryDeclerckRaquelFernandezGiacomo Ferrari Gil Francopoulo Dirk Heylen Julia HirschbergKristiinaJokinen MaciejKarpinski Staffan LarssonKiyong Lee OliverLemon CarlosMartinez-Hinarejos Paul Mc Kevitt Mike McTear David Novick Tim PaekPatriziaPaggio Catherine Pelachaud MassimoPoesioGermanRigauLaurentRomaryNiclaRossini Milan RuskoCandiceSidner Marieke van Erp Ielka van der Sluis KristinnThorissonAesoonYoonYorickWilks
ISO Linguistic Annotation Framework (Ide & Romary, 2005): annotations: the linguistic information that is added to segments of language data, independent of the format in which the information is represented; representations: particular formats in which annotations are rendered, (e.g. in XML, in typed feature structure AVMs, or in graphs in graphical form) independent of their content. Standards should be formulated at the level of annotations. ISO requirements for annotation standards
A concrete syntax defines an ideal representation format iff: Every annotation structure, defined by the abstract syntax, has a representation according the concrete syntax; Every representation, defined by the concrete syntax, is the rendering of a unique annotation structure according to the abstract syntax. Ideal Concrete Syntax
Ideal concrete syntax F1 abstract syntax ideal concrete syntax-1 F1-1 C21 F2-1 C12 F2 ideal concrete syntax-2 Ia semantics
An annotation structure is a pair <E, L> consisting of a set E of entity structures and set L of link structures, which connect entity structures. Entity structures contain semantic information about a segment of source data; link structures describe semantic relations between segments of source data. Most important type of entity structure in DiAML: dialogue act structure <S,A, d, f>, consisting of: speaker S; addressee A; dimension D; communicative function f or a pair <f, q> with qualifier(s) q DiAMLabstract syntax
1. P1: What time does the next train to Tilburg leave? Task: fs1: What time does the next train to Tilburg leave? Set-Question (‘WH-question’) 2. P2: The next train to Tilburg leaves I think at 8:32. Task: fs2: The next train to Tilburg leaves I think at 8:32. Answer [uncertain] AuFB: fs3: The next train to Tilburg Positive AutoFeedback DiAML example
Content of a context model / information state: All and exactly that information which has to be updated when interpreting a dialogue act. (And/or: all that information that is involved in the generation of dialogue acts) Formalization and implementation of a context model: DRSs (Poesio & Traum, 1998) ‘Contexts’ in Constructive Type Theory (Ahn, 2001) ‘Modular Partial Models’ (Bunt, 2002) ‘Dependent Record Types’ (Cooper, 2004) Typed Feature Structures (Keizer & Bunt, 2007) Common assumption: context models are highly structured, to facilitate efficient updating. Context models (information states)
Content of context models containing the kinds of information required by the 10 dimensions of DIT is best structured into 5 components to facilitate efficient representation and updating: Linguistic context: dialogue history; dialogue future Cognitive context: agent’s own processing of previous utterances; beliefs about partner’s processing Semantic context: information about the task (domain) Perceptual/physical context: perceptual information (e.g. visual) Social context: communicative obligations and permissions “Pending context”: information “under consideration”, which has yet to be evaluated for being consistent with content of “permanent” context. Context update = addition of elements to pending context. DIT context models
DIT context model implementation
Remember: dialogue act structure <S, A, d, f> . The (old) idea: the interpretation of a DA structure is a function that can be applied to a semantic content, resulting in the specification of an context update operation. Implementation: recursive valuation function V, recursion ending at application of model assignment F: V(<S, A, d, f> = V(f)(<F(S), F(A), F(d)>) . Taxonomy structure Update specification structure Update specifications as combinations of elementary update schemes Formalization of informal definitions, e.g. “S wants to make p known to A” Want(S, Bel(A, p)) “S believes p is correct” Bel(S,p) Dialogue act semantics
Elementary update schemes: e.g. U1: add to the addressee’s pending context that the speaker believes that p Example: V(<User, System, Task, PropositionalQuestion> = = V(PropQ)(<Usr, Sys, SemContext>) = = λp. [U10(Usr, Sys, Sys*SemC, p) υU11(Usr, Sys, Sys*SemC, p)] . e.g. if p= ArrTime(KL476, 19:15) then: (Sys* is Sys’ pending context) . U10(Usr, Sys, Sys*SemC, p) update: Sys believes that Usrwants to know whether ArrTime(KL476, 19:15); U11(Usr, Sys, Sys*SemC, p) update: Sys believes that Usrassumes that Sys knows whether ArrTime(KL476, 19:15) Dialogue act semantics (cont’d)
B: (u1) A: Could you please repeat that? F(Request) = λC. λX. λY. λD. λα. [U23(X, Y, D, α, C) υU26(X, Y, D, α, C)] Unconditional request: C = Τ (the universally true statement) F(Request)(A, B, Autofeedback, Repeat(u1), Τ) = = U23(A, B, CogContext, Repeat(u1), Τ) υU26(A, B, CogContext, Repeat(u1), Τ) = B*CogC=+ Bel(B, Want(A, [WillDo(B, Repeat(u1) -> CommitDo(B, Repeat(u1))]); B*CogC=+ Bel(B, Bel(A, CanDo(B, Repeat(u1)))) Example: feedback act with general-purpose function
A: Charlie? F(TurnAssign)(A, B) =[λX. λY. U101(X,Y,TurnM) υU102(X,Y, TurnM](A, B) = = U101(A,B,TurnM) υU102(X,Y,TurnM) = = B*LingC =+ Bel(B, Current-Speaker(A)) B*LingC=+ Bel(B, Want(A, Next-Speaker(B))) i.e. B believes that A currently has the speaker role; B believes that A wants B to next have the speaker role Example: Turn management act with dimension-specific function
Communicative function qualifiers: - q-specifiers: make the preconditions of the function that they qualify more specific, e.g. taking expression of (un-)certainty or (un-)conditionality into account. Semantics: V(< fi, qsj>) = fi‘(qsj’) - q-additives: enrich a communicative function with additional information, e.g.: with sentiment information V(< fi, qak>) = λS. λz. [fi’(S,z) υqak’(S,z)] Combination of q-specifier(s) and q-additive(s): V(< fi, qsj, qak>) = λS. λz. [(fi’(qsj’))(S,z) υ(qak’(S,z)] Semantics of qualifiers
A: How about a cup of coffee? B: That would be wonderful ! V(<AcceptOffer, unconditional, happy>) = = λS. λz. [[F(AcceptOffer)(F(unconditional))](S,z) υ[F(happy)](S,z)] = λS. λz. [[[λC. λX. λY. λD. λα. [[U24(X,Y,D_i, α) υU25(X,Y,D, α, C) υU25b(X,Y,D_i, α, C)](Τ)](S,z) υHAPPY(S,z)) = λS. λY. λD. λz.U24(S,Y,D, z) υU25(S,Y,D_i, z, Τ) υU25b(S,Y,D, z, Τ)] υHAPPY(S,z) Applied to A, B, and the action ‘coffee’: A*Task =+ Bel(A, Want(B, CommitDo(A, coffee))) A*Task =+ Bel(A, Bel(B, WillDo(A, coffee))) A*Task =+ Bel(A, HAPPY(B, coffee)) Semantics of qualifiers: example
Dialogue acts consist (essentially) of a communicative function and a semantic content. Communicative functions can be given a formal semantics in terms of updates of a context model. The semantic of communicative functions can be effectively implemented using a structured context model implementation, for example as a typed feature structure, and combinations of elementary update schemes. This works well for the fine-grained sets of ISO 24617-2 and DIT++ communicative functions, which distinguish around 90 ‘basic’ functions plus specifying and additive qualifiers for making more fine-grained distinctions. These communicative functions can effectively be recognized automatically (Petukhova & Bunt,IWCS 2011 paper) and open the perspective of effective full-blown incremental semantic interpretation of dialogue acts when integrated with incremental construction of a representation of the semantic content. Conclusions and Perspectives
References Bunt, Harry (2000)Dialoguepragmatics and context specification. In Harry Buntand William Black (eds.),Abduction, Belief and Context in Dialogue. Studies in ComputationalPragmatics, pp. 81-150. Amsterdam: John Benjamins. Bunt, Harry (2009) Multifunctionality and muldimensionaldialoguesemantics. In Proceedings of DiaHolmia, 13th Workshop on the Semanticsand Pragmatics of Dialogue, Stockholm, pp. 3-14. Bunt, Harry (2010) A methodologyfordesigningsemanticannotationlanguages. In Proceedings of the 2nd International Conference on Global InteroperabilityforLanguage Resources (ICGL-2), Hong Kong, January 2010, pp. 29-46. Bunt, Harry (in press) Interpretationand generation of dialoguewithmultidimensionalcontext models. In Anna Esposito (ed.), TowardAutonomous, adaptive, and context-aware multimedia interfaces, pp. 81--131. Berlin: Springer. Bunt, Harry (2011) Multifunctionality in dialogue. Computer, Speech and Language 25, 225-245. Bunt, Harry, Jan Alexandersson, Jean Carletta,Jae-WoongChoe, Alex Fang, KoitiHasida,Kiyong Lee,VolhaPetukhova,AndreiPopescu-Belis,LaurentRomary, Claudia Soria, and David Traum(2010). Towardsan ISO standardfordialogue actannotation. In Proceedings 7th International Conference onLanguageResources and Evaluation (LREC 2010). Paris: ELRA. Ide, Nancy and Harry Bunt (2010) Anatomyof semanticannotationschemes: Mappings to GrAF. In Proceedings of the 4th LinguisticAnnotation Workshop (LAW-IV), Uppsala. Poesio, Massimoand David Traum(1998) Towardsanaxiomatisation of dialogue acts. In Proceedingsof the Twente Workshop on the Semanticsand Pragmatics of Dialogue, Enschede, pp. 207-222.