We present an implementation of a discourse parsing system for a lexicalized Tree-Ajoining Grammar for discourse, specifying the integration of sentence and discourse level processing. Our system is based on the assumption that the compositional aspects of semantics at the discourse-level parallel those at the sentence-level. This coupling is achieved by factoring away inferential semantics and anaphoric features of discourse connectives. Computationally, this parallelism is achieved because both the sentence and discourse grammar are LTAG-based and the same parser works at both levels. The approach to an LTAG for discourse has been developed by Webber et al. in some recent papers ([33], [35], among others). Our system takes a discourse as input, parses the sentences individually, extracts the basic discourse consituent units from the sentence derivations, and reparses the discourse with reference to the discourse grammar while using the same parser used at the sentence-level.

Abstract not found

Abstract not found

Ordering information is a difficult but important task for applications generating natural-language text. We present a bottom-up approach to arranging sentences extracted for multi-document summarization. To capture the association and order of two textual segments (eg, sentences), we define four criteria, chronology, topical-closeness, precedence, and succession. These criteria are integrated into a criterion by a supervised learning approach. We repeatedly concatenate two textual segments into one segment based on the criterion until we obtain the overall segment with all sentences arranged. Our experimental results show a significant improvement over existing sentence ordering strategies. 1

D-LTAG is a discourse-level extension of lexicalized tree-adjoining grammar (LTAG), in which discourse syntax is projected by different types of discourse connectives and discourse interpretation is a product of compositional rules, anaphora resolution, and inference. In this paper, we present a D-LTAG extension of ongoing work on an LTAG syntax-semantic interface. First, we show how predicate-argument semantics are computed for standard, ‘structural ’ discourse connectives. These are connectives that retrieve their semantic arguments from their D-LTAG syntactic tree. Then we focus on discourse connectives that occur syntactically as (usually) fronted adverbials. These connectives do not retrieve both their semantic arguments from a single D-LTAG syntactic tree. Rather, their predicate-argument structure and interpretation distinguish them from structural connectives as well as from other adverbials that do not function as discourse connectives. The unique contribution of this paper lies in showing how compositional rules and anaphora resolution interact within the D-LTAG syntaxsemantic interface to yield their semantic interpretations, with multi-component syntactic trees sometimes being required. 1

Abstract. This paper describes a pedagogical agent that helps children to learn to author structured presentations about explanations of concepts. Using a Rhetorical Structure Theory analysis of a source Web page, the agent performs pedagogical tasks to support the user's understanding of rhetorical relations, stimulates reflection about the relations between the structure of the original text and the structure of the presentations, and suggests ways to improve the user's performance. Upon completion of the authoring, the presentations are organized into coherent structures that can be performed by animated characters, or Digital Puppets, in a learning-by-teaching classroom context. 1.

Abstract. In this paper, we describe a method of automatically generating summary slides from a text. The slides are generated by itemizing topic/non-topic parts that are extracted from the text based on syntactic/case analysis. The indentations of the items are controlled according to the discourse structure, which is detected by cue phrases, identification of word chain and similarity between two sentences. Our experiments demonstrates generated slides are far easier to read in comparison with original texts. 1

We use textual features motivated by systemic functional linguistic theory for genre-based text categorization. We have developed feature sets representing different types of conjunctions and modal assessment, which together indicate (partially) how different genres structure texts and express attitudes towards propositions in the text. Using such features enables analysis of large-scale rhetorical differences between genres by examining which features are important for classification. The specific domain studied comprises scientific articles in historical and experimental sciences (paleontology and physical chemistry respectively). The SMO learning algorithm with our feature set achieved over 83 % accuracy for classifying articles according to field, though no field-specific terms were used as features. The most highly-weighted features were consistent with hypothesized methodological differences between historical and experimental sciences, thus lending empirical evidence to the notion of multiple scientific methods.

Over the past decade, Question Answering (QA) has generated considerable interest and participation in the fields of Natural Language Processing and Information Retrieval. Conferences such as TREC, CLEF and DUC have examined various aspects of the QA task in the academic community. In the commercial world, major search engines from Google, Microsoft and Yahoo have integrated basic QA capabilities into their core web search. These efforts have focused largely on so-called “factoid ” questions seeking a single fact, such as the birthdate of an individual or the capital city of a country. Yet in the past few years, there has been growing recognition of a broad class of “long-answer ” questions which cannot be satisfactorily answered in this framework, such as those seeking a definition, explanation, or other descriptive information in response. In this thesis, we consider the problem of answering such questions, with particular focus on the contribution to be made by integrating rhetorical and semantic models. We present DefScriber, a system for answering definitional (“What is X?”), biographi-cal (“Who is X?”) and other long-answer questions using a hybrid of goal- and data-driven methods. Our goal-driven, or top-down, approach is motivated by a set of definitional pred-

We are compiling a corpus of Dutch texts annotated with discourse structure and lexical cohesion, containing initially 80 texts from expository and persuasive genres. We are using this resource for corpus-based studies of discourse relations, discourse markers, cohesion, and genre differences. We are also exploring the possibilities of automatic text segmentation and semi-automatic discourse annotation. This paper discusses our design choices in text selection and segmentation and in the annotation of discourse structure and lexical cohesion. 1