this paper, we focus on two levels of analysis. The first involves the relation between the information conveyed in consecutive elements of a coherent discourse. Thus, for example, one utterance may describe an event that can be presumed to be the cause of another event described in the subsequent utterance. This causal relation is at what we will call the informational level. The second level of relation results from the fact that discourses are produced to effect changes in the mental state of the discourse participants. In coherent discourse, a speaker is carrying out a consistent plan to achieve the intended changes, and consecutive discourse elements are related to one another by means of the ways in which they participate in that plan. Thus, one utterance may be intended to increase the likelihood that the hearer will come to * Department of Computer Science and Intelligent Systems Program, University of Pittsburgh, Pittsburgh, PA 15260. e-mail: jmoore,pollackcs.pitt.edu 1 In addition, intentional structure is needed to make certain types of choices during the generation process, e.g., how to refer to an object (Appelt 1985)

We describe experiments that show that the concepts of rhetorical analysis and nuclearity can be used effectively for determining the most important units in a text. We show how these concepts can be implemented and we discuss results that we obtained with a discourse-based summarization program. 1 Motivation The evaluation of automatic summarizers has always been a thorny problem: most papers on summarization describe the approach that they use and give some "convincing " samples of the output. In very few cases, the direct output of a summarization program is compared with a human-made summary or evaluated with the help of human subjects; usually, the results are modest. Unfortunately, evaluating the results of a particular implementation does not enable one to determine what part of the failure is due to the implementation itself and what part to its underlying assumptions. The position that we take in this paper is that, in order to build high-quality summarization programs, one ...

This thesis is an inquiry into the nature of the high-level, rhetorical structure of unrestricted natural language texts, computational means to enable its derivation, and two applications (in automatic summarization and natural language generation) that follow from the ability to build such structures automatically. The thesis proposes a first-order formalization of the high-level, rhetorical structure of text. The formalization assumes that text can be sequenced into elementary units; that discourse relations hold between textual units of various sizes; that some textual units are more important to the writer's purpose than others; and that trees are a good approximation of the abstract structure of text. The formalization also introduces a linguistically motivated compositionality criterion, which is shown to hold for the text structures that are valid. The thesis proposes, analyzes theoretically, and compares empirically four algorithms for determining the valid text structures of ...

Abstract not found

This paper addresses this issue in the context of the expression of procedural relations between actions in instructional text. It employs the following four step approach to achieve this goal: (1) Collect a corpus of the relevant text type; (2) Perform a detailed linguistic study of a portion of this corpus, called the training set, and reserving the remainder as a testing set; (3) Implement the results of this study in a text generation system; (4) Compare the output of the system with the text found in the entire corpus. This has resulted in the construction of IMAGENE, an instructional text generation system which embodies a model of the forms of expression consistently used by instructional text writers over a broad range of instruction types. The details of IMAGENE's

In order to provide a principled foundation for the study of discourse, in this paper we propose answers to three basic questions: What are the atomic units of discourse? What kind of structures can be built from the elementary units? How do we interpret the resulting structures semantically? Inferences and the correct interpretation of deixis and anaphors in discourse depend upon both structural and semantic accessibility relations. Structurally, we argue, discourse is context free and accessibility is determined by the coordination and subordination relations specified by the model of discourse presented here. Semantically, accessibility is controlled by relations among a number of modal contexts (interaction, speech event, genre unit, modality, polarity, and point of view) which determine the discourse world relative to which each primitive discourse unit is interpreted. To demonstrate the validity of our approach, the linguistic discourse model developed here is applied to a problem concerning the distribution of a discourse particle in Mocho and to various problems of discourse interpretation.

We perform a linguistic analysis of documents during indexing for information retrieval. By eliminating index terms that occur only in subordinate clauses, index size is reduced by approximately 30% without adversely affecting precision or recall. These results hold for two corpora: a sample of the world wide web and an electronic encyclopedia.

RASTA (Rhetorical Structure Theory Analyzer), a system for automatic discourse analysis, reliably identifies rhetorical relations present in written discourse by examining information available in syntactic and logical form analyses. Since there is a many-to-many relationship between rhetorical relations and elements of linguistic form, RASTA identifies relations by the convergence of a number of pieces of evidence, many of which would be insufficient in isolation to reliably identify a relation. 1. Introduction Within Rhetorical Structure Theory (RST) (Mann and Thompson 1986, 1988), the discourse structure of a text is represented by means of a hierarchical tree diagram in which contiguous text spans are related by labeled relations. Hierarchical structure results from the fact that each text span in a labeled relation may itself have a complex internal discourse structure. Traditionally, human analysts have constructed RST analyses for texts by employing tacit, subjective, intuitiv...

The term discourse structure is used to denote any structure of a text above that of the sentence. Trees have often been posited as a good abstraction when discourse is taken to have a hierarchical structure (Mann and Thompson 1987; Webber et al. 2003; Marcu 2000; Egg and Redeker 2008). Nevertheless, periodically researchers have commented on the need to depart from the strict singleparent hierarchy of trees to structures which have shared daughters, a move which incorporates multiple inheritance and is therefore an issue for tree representations. This study follows up on the observation in (Lee et al. 2006) about the relative ubiquity of shared structures in the Penn Discourse Treebank or PDTB (Prasad et al. 2008; PDTB-Group 2008)), a recently released corpus which annotates discourse relations and their arguments. We limit our investigation here to cases where the shared discourse structure is a syntactically subordinate clause introduced by a subordinating conjunction (e.g. because, although, when, etc.). We examine annotations in the PDTB where the subordinate clause has been taken to be an argument of both the relation associated with the subordinating conjunction and another relation expressed in the immediately subsequent discourse. We ask what such annotations imply about the link between syntactic subordination and discourse subordination. Our argument is that while syntactic subordination may often correlate with discourse subordination, there are interesting exceptions that might better be captured as discourse coordination. We provide some systematic characterization of these exceptions by appealing to well-motivated discourse factors, and discuss their implications for tree structures. 1

Introduction The most common assumption about intentiou in discourse is that tile primary intention of discourse is to comlnunicate and receive information. This is a founding assumption of every formal model of discourse meaning that I am aware of. The standard account of meaning is that utterances are functions from contexts to contexts whose primary purpose is to describe the world, aud whose meaning derives from the fact that they delimit the set of worlds that the conversants believe possible. One of the ramifications of this assumption is that utterances with no new information are infelicitous or have no meaning[1, 2, 14, 5]. However, consider example l, asserted by a passenger in a vehicle in response to tile driver's comment that the heavy traffic wins unexpected: (I) There's something on fire up there. I can't see what's on fire, but SOMETHING IS. (LW 6/12/92) In the first clause of 1, the speaker asserts a proposition P, namely that something is on fire. In the second clau