Showing papers by "Nicholas Asher published in 2010"

Testing SDRT's Right Frontier

Abstract: The Right Frontier Constraint (RFC), as a constraint on the attachment of new constituents to an existing discourse structure, has important implications for the interpretation of anaphoric elements in discourse and for Machine Learning (ML) approaches to learning discourse structures. In this paper we provide strong empirical support for SDRT's version of RFC. The analysis of about 100 doubly annotated documents by five different naive annotators shows that SDRT's RFC is respected about 95% of the time. The qualitative analysis of presumed violations that we have performed shows that they are either click-errors or structural misconceptions.

SDRT and continuation semantics

Abstract: Segmented Discourse Representation Theory (SDRT) [2,7] provides a dynamic semantics for discourse that exploits a rich notion of discourse structure. According to SDRT, a text is segmented into constituents related to each other by means of rhetorical relations; the resulting structure, known as a segmented discourse representation structure or SDRS has various semantic effects. This theory has shown how discourse structure makes contributions to the interpretation of a variety of linguistic phenomena, including tense, modality, presupposition, the interpretation of anaphoric pronouns and ellipsis. SDRT exploits dynamic semantics [20,14] to interpret SDRSs. We investigate here the advantages of integrating SDRT within continuation style semantics of the sort developed in [17].

Testing SDRT's Right Frontier

Abstract: The Right Frontier Constraint (RFC), as a constraint on the attachment of new constituents to an existing discourse structure, has important implications for the interpretation of anaphoric elements in discourse and for Machine Learning (ML) approaches to learning discourse structures. In this paper we provide strong empirical support for SDRT's version of RFC. The analysis of about 100 doubly annotated documents by five different naive annotators shows that SDRT's RFC is respected about 95% of the time. The qualitative analysis of presumed violations that we have performed shows that they are either click-errors or structural misconceptions.

A Montagovian Treatment of Modal Subordination

Abstract: Epistemic modality involves complex contextual dependencies that linguists have studied (Roberts 1989; Veltman 1996) We show that they have a natural treatment within a continuation style semantics

Biased questions, intonation, and discourse

Abstract: Semanticists have demonstrated repeatedly over the past twenty years that intonation often conveys information important for determining the content of a discourse. For example, intonation is important for marking focus in English, which in turn is important for interpreting sentences with focus sensitive adverbs like even and only (Beaver and Clark 2003). Intonation is also important in marking the discourse function of utterances in discourse and dialogue. For example, it is an essential clue in determining whether or not an assertion counts as a felicitous answer to a question given in the prior discourse context. The canonical way of presenting an answer to a question such as (1a) is to place the nuclear pitch accent on the constituent that replaces the wh-particle, as in (1b) (Kadmon 2001). Alternative realizations of the same sentence are anomalous, as shown in (1c).

Extracting and modelling preferences from dialogue

Abstract: Dialogue moves influence and are influenced by the agents' preferences. We propose a method for modelling this interaction. We motivate and describe a recursive method for calculating the preferences that are expressed, sometimes indirectly, through the speech acts performed. These yield partial CP-nets, which provide a compact and efficient method for computing how preferences influence each other. Our study of 100 dialogues in the Verbmobil corpus can be seen as a partial vindication of using CP-nets to represent preferences.