Framework for a model of dialogue (original) (raw)

The structure of dialog

1999

The purpose of this paper is to provide a kind of tutorial concerning some of the phenomena which contribute to the structure of dialog. It, therefore, gives an overview rather than a detailed account of these phenomena. It must also be said that the overview is given from a linguistic pragmatic perspective, namely, that of" activity based communication analysis" and summarizes some of the more detailed accounts published elsewhere, cf. for example Allwood, 1995.

Modelling dialogue as inter-action

International Journal of Speech Technology, 2008

In this paper we introduce a formal model of dialogue based on grammar systems theory: Conversational Grammar Systems (CGS). The model takes into account ideas from the study of human-human dialogue in order to define a flexible mechanism for coherent dialogues that may help in the design of effective and user-friendly computer dialogue systems. The main feature of the model is to present an action view of dialogue. CGS model dialogue as an inter-action, this is a sequence of acts performed by two or more agents in a common environment. We claim that CGS are able to model dialogue with a high degree of flexibility, which means that they are able to accept new concepts and modify rules, protocols and settings during the computation.

A Model of Natural Language Dialogue

1993

In this thesis I have a threefold purpose. Firstly, I will attempt to argue that the individual utterances agents make in natural language dialogue stem from specific beliefs, goals, and plans and that these interlock with those of other agents in the production of dialogue. I suggest that agents utilise syntactic, semantic, pragmatic and contextual knowledge in this process. Furthermore, that these elements contribute to the utterances speakers make and hearers interpret in the pursuit of their individual goals, and cannot be treated separately. I will suggest that utterances, being intentional behaviour, are sub-components of plans to achieve specific communicative purposes. Following from this, I will present a descriptive model showing how the beliefs and goals of agents contribute to the composing of a logical form for an utterance prior to its syntactic representation. It is suggested that the logical form of an utterance, is composed of elements relating to the agent's be...

Non-sentential utterances in dialogue: classification, resolution and use

2006

This thesis combines corpus work with symbolic and statistical techniques to offer an account of non-sentential utterances in dialogue. These are utterances that do not have the form of a full sentence according to most traditional grammars, but that nevertheless convey a complete sentential meaning. The approach taken analyses non-sentential utterances as radical context-dependent expressions, and formalises them as functions from the current information state to the type of the next information state. The thesis is grounded in a corpus study that provides evidence about the occurrence of non-sentential utterances in conversation. This is used to put forward a data-driven taxonomy of classes of non-sentential utterances, which are then paired with a formal specification of their resolution requirements formulated in Type Theory with Records. In order to identify the right class, which will determine the appropriate resolution procedure, machine learning methods are employed, which can be used to boost the automatic processing of dialogue. The thesis also proposes a hierarchy of models for dialogue protocols, defined in terms of abstract machine models, that is based on a variety of structural features of dialogue that are also related to properties of non-sentential utterances. This suggest a link between dialogue dynamics on the one hand, and formal language theory and the theory of computation in the other.

La représentation des actes de langage pour traiter le dialogue

Actes De L Atelier Thematique Theories Semantiques Et Pragmatiques Le Temps L Espace Et Le Mouvement Du Lexique Au Discours Et Au Dialogue Taln 99, 1999

In this paper, we examine the way in which formal semantics such as SDRT (Asher, 1993) treat dialogue. We point out that SDRT offers two different solutions to combine utterances in a dialogue, one based on reported speech and one based on new discourse relations specific to sequences of utterances . We show how this can be problematic especially for access to discourse referents. We think that even at a semantic level, treating dialogue cannot be done in a context-free way. Specifically, we show how representing events of utterance production allows for an homogeneous treatment of both dialogue and its representation in reported speech. We claim that such an approach allows accounting for the structure of dialogue and show the advantages which it offers for the treatment of reference, especially where deictics are concerned.

Prosodic and lexical indications of discourse structure in human-machine interactions

Speech Communication, 1997

From a discourse perspective, utterances may vary in at least two important respects: (i) they can occupy a different hierarchical position in a larger-scale information unit and (ii) they can represent different types of speech acts. Spoken language systems will improve if they adequately take into account both discourse segmentation and utterance purpose. An important question then is how such discourse-structural features can be detected. Analyses of monologues and human-human dialogues have shown that a good indicator of these factors is prosody, defined as the set of suprasegmental speech features. This paper explores whether speakers also use prosody to highlight discourse structure in a particular type of human-machine interaction, viz., information query in a travel-planning domain. More specifically, it investigates if speakers signal (i) the start of a new topic by marking the initial utterance of a discourse segment, and (ii) whether an utterance is a normal request for information or part of a correction sub-dialogue. The study reveals that in human-machine interactions, both discourse segmentation and utterance purpose can have particular prosodic correlates, although speakers also mark this information through choice of wording. Therefore, it is useful to explore in the future the possibilities of incorporating prosody in spoken language systems as a cue to discourse structure. 0 1997 Elsevier Science B.V.

ON MODELINGThe language/action model of conversation

Interactions, 2010

Editor's Note: In last year's January + February issue Usman Haque, Paul Pangaro, and I described several types of interaction---reacting, regulating, learning, balancing, managing, and conversing. In the July + August 2009 issue, Paul Pangaro and I described several types of conversing---agreeing, learning, coordinating, and collaborating---and we proposed using models based on Gordon Pask's Conversation Theory as a guide for improving human-computer interaction. Peter Jones responded, noting that there are other models of conversation and prior work in bringing conversation to human-computer interaction in particular Winograd and Flores 1986 work with The Coordinator. We agree on the importance of The Coordinator and invited Peter to outline the history of models of conversation and their relationship to HCI. His response follows.---Hugh Dubberly

Conversational actions and discourse situations

Computational intelligence, 1997

We use the idea that actions performed in a conversation become part of the common ground as the basis for a model of context that reconciles in a general and systematic fashion the differences between the theories of discourse context used for reference resolution, intention recognition, and dialogue management. We start from the treatment of anaphoric accessibility developed in discourse representation theory (DRT), and we show first how to obtain a discourse model that, while preserving DRT's basic ideas about referential accessibility, includes information about the occurrence of speech acts and their relations. Next, we show how the different kinds of 'structure' that play a role in conversation-discourse segmentation, turn-taking, and grounding-can be formulated in terms of information about speech acts, and use this same information as the basis for a model of the interpretation of fragmentary input.