RST Discourse Treebank Research Articles

Continuity in discourse relations

Abstract Continuity and discontinuity (maintaining or shifting deictic centres across segments) are important aspects of discourse relations. Yet they have been attributed to these relations in very different ways. This calls for an analysis of individual instances of discourse relations with respect to their continuity dimensions. To this end, we operationalise Givón’s (1993) continuity dimensions (time, space, reference, action, perspective, modality, and speech act), decomposing them into distinctive features that allow a consistent and accurate classification of the continuity dimensions in discourse relation tokens. This inventory was applied to five representative relation types (causal, contrastive, conditional, elaboration, and temporal) from the RST Discourse Treebank (Carlson & Marcu 2001). We found that relations can simultaneously be more continuous for some dimensions but more discontinuous for others. What is more, discourse relations typically vary widely in different continuity dimensions and thus cannot be described as fully continuous or discontinuous, neither on the level of the entire relation type nor for one of its particular dimensions. Using examples of causal, conditional, and contrastive relations, we also illustrate how the results of our analysis can be used to verify hypotheses about correlations between continuity and discourse relations.

Signalling conditional relations

Abstract We investigate how discourse relations and their subtypes are signalled, extending the set of discourse signals from connectives and lexical cue phrases to the wide range of semantic, syntactic, and orthographic signals of the RST Signalling Corpus (Das, Debopam & Maite Taboada. 2018. RST signalling corpus. Language Resources and Evaluation 52. 149–184). This extension requires re-evaluating previous predictions on discourse signalling, in particular, those of Sanders, Ted. 2005. Coherence, causality and cognitive complexity in discourse. In M. Aurnague, M. Bras, A. Le Draoulec & L. Vieu (eds.), Proceedings/Actes SEM-05, first international symposium on the exploration and modelling of meaning, 105–114. Biarritz causality-by-default hypothesis, the hypothesis of uniform information density (Frank, Austin & Florian Jaeger. 2008. Speaking rationally: Uniform information density as an optimal strategy for language production. In Proceedings of the 30th annual meeting of the Cognitive Science Society, 933–938. https://escholarship.org/uc/item/7d08h6j4 (accessed 18 May 2022)), and the hypothesis that discourse is continuous by preference (Segal, Erwin, Judith Duchan & Paula Scott. 1991. The role of interclausal connectives in narrative structuring. Discourse Processes 14. 27–54; Murray, John. 1997. Connectives and narrative text. Memory and Cognition 25. 227–236). We evaluate the predictions of these theories on the conditional relations in the RST Discourse Treebank (Carlson, Lynn, Daniel Marcu & Mary Ellen Okurowski. 2002. RST Discourse Treebank. LDC2002T07. Philadelphia: Linguistic Data Consortium), using causal relations as a control group. Informativity and continuity are operationalized in terms of semantic complexity and Givón, Talmy. 1993. English grammar: A function-based introduction, vol. 2. Amsterdam: John Benjamins dimensions of deictic shift. Our results show that the hypotheses make accurate predictions only for the relation groups in their entirety but not for the observed in-group variation, in particular, the low amount of marking for the hypothetical subtype of conditional relations. We attribute this difference to the distribution of intra- and inter-sentential occurrences across the conditional subtypes: intra-sentential relations are consistently more marked than inter-sentential ones, and hypothetical relations are special in that they appear predominantly inter-sententially.

A computational model for measuring discourse complexity

In past studies, the few quantitative approaches to discourse structure were mostly confined to the presentation of the frequency of discourse relations. However, quantitative approaches should take into account both hierarchical and relational layers in the discourse structure. This study considers these factors and addresses the issue of how discourse relations and discourse units are related. It draws upon the available corpora of discourse structure (rhetorical structure theory-discourse treebank (RST-DT)) from a new perspective. Since an RST tree can be converted into a syntactic dependency tree, the data extracted from the RST-DT can be useful for calculating the discourse distance in much the same way as syntactic dependency distance is calculated. Discourse distance is also applicable to measuring the depth of the human processing of discourse. Furthermore, the data derived from the RST-DT are also easily converted into network data. This study finds that discourse structure has its discourse distance minimum and each type of RST relations has its range of discourse distance. The frequency distribution of discourse data basically follows the power law on several levels, while a network approach reveals how discourse units are arranged spatially in regular patterns. The two methods are mutually complementary in revealing the interaction between discourse relations and discourse units in a comprehensive manner, as well as in revealing how people process and comprehend discourse dynamically. Accordingly, we propose merging the two methods so as to yield a computational model for assessing discourse complexity and comprehension.

How compatible are our discourse annotation frameworks? Insights from mapping RST-DT and PDTB annotations

Discourse-annotated corpora are an important resource for the community, but they are often annotated according to different frameworks. This makes joint usage of the annotations difficult, preventing researchers from searching the corpora in a unified way, or using all annotated data jointly to train computational systems. Several theoretical proposals have recently been made for mapping the relational labels of different frameworks to each other, but these proposals have so far not been validated against existing annotations. The two largest discourse relation annotated resources, the Penn Discourse Treebank and the Rhetorical Structure Theory Discourse Treebank, have however been annotated on the same texts, allowing for a direct comparison of the annotation layers. We propose a method for automatically aligning the discourse segments, and then evaluate existing mapping proposals by comparing the empirically observed against the proposed mappings. Our analysis highlights the influence of segmentation on subsequent discourse relation labelling, and shows that while agreement between frameworks is reasonable for explicit relations, agreement on implicit relations is low. We identify several sources of systematic discrepancies between the two annotation schemes and discuss consequences for future annotation and for usage of the existing resources.

A Dependency Perspective on RST Discourse Parsing and Evaluation

Computational text-level discourse analysis mostly happens within Rhetorical Structure Theory (RST), whose structures have classically been presented as constituency trees, and relies on data from the RST Discourse Treebank (RST-DT); as a result, the RST discourse parsing community has largely borrowed from the syntactic constituency parsing community. The standard evaluation procedure for RST discourse parsers is thus a simplified variant of PARSEVAL, and most RST discourse parsers use techniques that originated in syntactic constituency parsing. In this article, we isolate a number of conceptual and computational problems with the constituency hypothesis. We then examine the consequences, for the implementation and evaluation of RST discourse parsers, of adopting a dependency perspective on RST structures, a view advocated so far only by a few approaches to discourse parsing. While doing that, we show the importance of the notion of headedness of RST structures. We analyze RST discourse parsing as dependency parsing by adapting to RST a recent proposal in syntactic parsing that relies on head-ordered dependency trees, a representation isomorphic to headed constituency trees. We show how to convert the original trees from the RST corpus, RST-DT, and their binarized versions used by all existing RST parsers to head-ordered dependency trees. We also propose a way to convert existing simple dependency parser output to constituent trees. This allows us to evaluate and to compare approaches from both constituent-based and dependency-based perspectives in a unified framework, using constituency and dependency metrics. We thus propose an evaluation framework to compare extant approaches easily and uniformly, something the RST parsing community has lacked up to now. We can also compare parsers’ predictions to each other across frameworks. This allows us to characterize families of parsing strategies across the different frameworks, in particular with respect to the notion of headedness. Our experiments provide evidence for the conceptual similarities between dependency parsers and shift-reduce constituency parsers, and confirm that dependency parsing constitutes a viable approach to RST discourse parsing.

Quantitative Aspects of PDTB-Style Discourse Relations across Languages

Frequency distribution of words, syntax and semantics in many languages abides by certain laws. However, because of the shortage of discourse corpora, few studies have examined whether the frequency of discourse relations follows some distributional patterns. Although there is some research based on the Rhetorical Structure Theory discourse treebank (RST-DT), each of these studies is limited to a single language. Otherwise to the RST-DT, the Penn Discourse Treebank (PDTB), adopting another annotation system, has had an enormous influence on the study of discourse structure and discourse annotation. Discourse corpora in other languages, such as Chinese, Hindi, Turkish, Czech and Arabic have been annotated following PDTB style. With the data from these discourse treebanks, we find that the rank-frequency of discourse relations follow the same pattern and that these languages share significant similarities in using semantic relations to organize the discourse. It is evidenced in our research that humans assume the relationship between two consecutive sentences is a causal connection or expansion link for fewer connectives used, but the relation of contrast is the most marked by connectives. This research will be of significance for understanding the homogeneity of discourse structure across languages.

Signalling of Coherence Relations in Discourse, Beyond Discourse Markers

ABSTRACTWe argue that coherence relations (relations between propositions, such as Concession or Purpose) are signalled more frequently and by more means than is generally believed. We examine how coherence relations in text are indicated by all possible textual signals, and whether every relation is signalled. To that end, we conducted a corpus study on the RST Discourse Treebank, a corpus of newspaper articles annotated for rhetorical (or coherence) relations. Results from our corpus study show that most relations in text (over 90%) are signalled and also that most signalled relations (over 80%) are indicated not only by discourse markers (and, but, if, since), but also by a wide variety of signals other than discourse markers, such as reference, lexical, semantic, syntactic and graphical features. These findings suggest that signalling of coherence relations is much more sophisticated than previously thought.

A CDT-Styled End-to-End Chinese Discourse Parser

Discourse parsing is a challenging task and plays a critical role in discourse analysis. Since the release of the Rhetorical Structure Theory Discourse Treebank and the Penn Discourse Treebank, the research on English discourse parsing has attracted increasing attention and achieved considerable success in recent years. At the same time, some preliminary research on certain subtasks about discourse parsing for other languages, such as Chinese, has been conducted. In this article, we present an end-to-end Chinese discourse parser with the Connective-Driven Dependency Tree scheme, which consists of multiple components in a pipeline architecture, such as the elementary discourse unit (EDU) detector, discourse relation recognizer, discourse parse tree generator, and attribution labeler. In particular, the attribution labeler determines two attributions (i.e., sense and centering) for every nonterminal node (i.e., discourse relation) in the discourse parse trees. Systematically, our parser detects all EDUs in a free text, generates the discourse parse tree in a bottom-up way, and determines the sense and centering attributions for all nonterminal nodes by traversing the discourse parse tree. Comprehensive evaluation on the Connective-Driven Dependency Treebank corpus from both component-wise and error-cascading perspectives is conducted to illustrate how each component performs in isolation, and how the pipeline performs with error propagation. Finally, it shows that our end-to-end Chinese discourse parser achieves an overall F1 score of 20% with full automation.

RST Signalling Corpus: a corpus of signals of coherence relations

We present the RST Signalling Corpus (Das et al. in RST signalling corpus, LDC2015T10. https://catalog.ldc.upenn.edu/LDC2015T10 , 2015), a corpus annotated for signals of coherence relations. The corpus is developed over the RST Discourse Treebank (Carlson et al. in RST Discourse Treebank, LDC2002T07. https://catalog.ldc.upenn.edu/LDC2002T07 , 2002) which is annotated for coherence relations. In the RST Signalling Corpus, these relations are further annotated with signalling information. The corpus includes annotation not only for discourse markers which are considered to be the most typical (or sometimes the only type of) signals in discourse, but also for a wide array of other signals such as reference, lexical, semantic, syntactic, graphical and genre features as potential indicators of coherence relations. We describe the research underlying the development of the corpus and the annotation process, and provide details of the corpus. We also present the results of an inter-annotator agreement study, illustrating the validity and reproducibility of the annotation. The corpus is available through the Linguistic Data Consortium, and can be used to investigate the psycholinguistic mechanisms behind the interpretation of relations through signalling, and also to develop discourse-specific computational systems such as discourse parsing applications.

Motifs in Reconstructed RST Discourse Trees

In line with the compositionality criterion and hierarchy principle of Rhetorical Structure Theory (RST), this study converts each tree in the RST Discourse Treebank into three trees with mere ultimate nodes being clauses, sentences and paragraphs, respectively. It examines the motifs of rhetorical relations along three taxonomies at the three granularity levels and also lengths of these motifs, and finds they observe the negative binomial distribution and positive negative binomial distribution respectively. The study demonstrates the applicability of RST relational analysis between same-level terminal units, which works with various granularities.

Rhetorical relations revisited across distinct levels of discourse unit granularity

In accordance with the compositionality criterion and hierarchy principle of Rhetorical Structure Theory (RST), this study reframes each tree in the RST Discourse Treebank into three new dependency trees with ultimate nodes being clauses, sentences, and paragraphs, respectively, which also draw on an analogy between syntactic and discourse trees. Detailed percentages of various RST relations at the three granularity levels are examined, illuminating the discourse processes of organizing units of one granularity level into those of the next upper level and suggesting certain homogeneity and interaction across levels in the Treebank, particularly at the two upper levels. The study demonstrates the applicability of RST analysis between same-level terminal units. With unique analytical advantages, the newly constructed discourse dependency trees provide new research prospects.

Building a Language-Independent Discourse Parser using Universal Networking Language

Discourse parsing has become an inevitable task to process information in the natural language processing arena. Parsing complex discourse structures beyond the sentence level is a significant challenge. This article proposes a discourse parser that constructs rhetorical structure RS trees to identify such complex discourse structures. Unlike previous parsers that construct RS trees using lexical features, syntactic features and cue phrases, the proposed discourse parser constructs RS trees using high-level semantic features inherited from the Universal Networking Language UNL. The UNL also adds a language-independent quality to the parser, because the UNL represents texts in a language-independent manner. The parser uses a naive Bayes probabilistic classifier to label discourse relations. It has been tested using 500 Tamil-language documents and the Rhetorical Structure Theory Discourse Treebank, which comprises 21 English-language documents. The performance of the naive Bayes classifier has been compared with that of the support vector machine SVM classifier, which has been used in the earlier approaches to build a discourse parser. It is seen that the naive Bayes probabilistic classifier is better suited for discourse relation labeling when compared with the SVM classifier, in terms of training time, testing time, and accuracy.