Elhadad, M. (1995). "Using argumentation in text generation." Journal of Pragmatics 24: 189-220.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....as part of a complete argument generator. Other modules of the generator include a microplanner, which performs aggregation, pronominalization and makes decisions about cue phrases and scalar adjectives, along with a sentence realizer, which extends previous work on realizing evaluative statements (Elhadad 1995). 2.1 Background on AMVF An AMVF is a model of a person s values and preferences with respect to entities in a certain class. It comprises a value tree and a set of component value functions, one for each primitive attribute of the entity. A value tree is a decomposition of the value of an ....

Elhadad, M. (1995). "Using argumentation in text generation." Journal of Pragmatics 24: 189-220.

....In the field of natural language generation (NLG) considerable research has been devoted to develop computational models for automatically generating user tailored evaluative arguments. Among others, 1 3] have investigated the process of selecting and structuring the argument content, while [4] developed a detailed model of how the selected content should be realised into natural language. However, a key limitation of previous research is that specific projects have tended to focus on only one aspect of the generation process, leaving the development of a comprehensive computational ....

....the packaging of semantic information into sentences, is performed according to the standard techniques summarized in Section 2. With respect to lexicalization, the GEA microplanner selects words to express evaluations by following an extension of previous work on realizing evaluative statements [4], whereas decisions about cue phrases (to express discourse relationships among text segments) are implemented as a decision tree based on features suggested in the literature (e.g. 11] 12] The generation of referring expression in GEA is straightforward; an entity is always referred to by ....

Elhadad, M., Using argumentation in text generation. Journal of Pragmatics, 1995. 24: p. 189-220.

.... We have found no attempt to use a systematic approach for the definition of such abstract concepts as arguments , values or opinions , which, besides being interesting in their own right, are widely used, more or less explicitly, in dialogue or natural language generation computer systems [11, 17, 29, 32, 36, 37, 42], and therefore constitute a potential, important set of sharable concepts, in the ontology engineering community spirit. As to comments to our conceptualization process, the use of explicit guidelines, coming from an ontology conceptualization methodology, has actually proved to significantly ....

M. Elhadad, `Using Argumentation in Text Generation', Journal of Pragmatics, 24, 189--220, (1995).

.... Argumentative Strategy Our framework for generating evaluative arguments is based on previous work in artificial intelligence on explaining decision theoretic advice (Klein and Shortliffe, 1994) and on previous work in computational linguistics on generating natural language evaluative arguments (Elhadad, 1995) . On the one hand, the study on explaining decision theoretic advice produced a rich quantitative model that can serve as a basis for strategies to select and organize the content of decisiontheoretic explanations, but was not concerned at all with linguistic issues. On the other hand, the work ....

Elhadad, M. (1995). Using argumentation in text generation. Journal of Pragmatics 24:189--220.

.... a standard approach, and this is probably not surprising (how can we say that a text is good ) Can Argumentation Help As most, if not all, of the Argumentation process is based on the use language it is surprising that Argumentation theory and NLP meet only in relatively rare occasions (see [7, 8, 16, 18, 22, 24, 29] inter alia) and for addressing specific problems, which almost never go beyond the generation understanding of a single sentence or paragraph, typically in support or attacking a claim. An important, common misconception which pervades all these experiences, in my opinion, is that Argumentation ....

M. Elhadad. Using Argumentation in Text Generation. Journal of Pragmatics, 24:189--220, 1995.

.... Argumentative Strategy Our framework for generating evaluative arguments is based on previous work in artificial intelligence on explaining decision theoretic advice (Klein and Shortliffe, 1994) and on previous work in computational linguistics on generating natural language evaluative arguments (Elhadad, 1995) . On the one hand, the study on explaining decision theoretic advice produced a rich quantitative model that can serve as a basis for strategies to select and organize the content of decisiontheoretic explanations, but was not concerned at all with linguistic issues. On the other hand, the work ....

Elhadad, M. (1995). Using argumentation in text generation. Journal of Pragmatics 24:189--220.

....have approached the problem of presenting arguments to the user. Most of them are principally focused on the structure of valid arguments (Fox, Krause, Elvang Goransson 1993; Karacapidilis 1996; Vreeswijk 1997) Others are mainly interested in the generation of arguments in natural language (Elhadad 1995; Maybury 1993; Reed Long 1997b) Fewer are the cases in which the effectiveness of arguments is considered, as opposed to their validity: Zukerman, Korb, McConachy (1996) describe the architecture of a system capable of generating nice arguments tailored to the addressee, but the only ....

Elhadad, M. 1995. Using Argumentation in Text Generation. Journal of Pragmatics 24:189--220.

....to the addressee, but the only concession to validity here is allowing an argument to have some steps missing in its logic chain. Closer to our interests, other approaches in which the system is requested to produce arguments are mainly concerned with the generation of them in natural language [15, 26, 30], with the emphasis on a demonstrative style of arguing, aiming at changing the belief state of the addressee with respect to a proposition. The aims of an advisor, as Daphne is meant to be, are more concerned with values than with beliefs, and demonstrative arguments are only a subset of the ....

M. Elhadad, `Using Argumentation in Text Generation', Journal of Pragmatics, 24, 189--220, (1995).

....every point where the user is supposed to intervene, a menu of possible replies is proposed, among which the user will choose the most appropriate. The formalization differs from other theoretical or applicative works in argumentation (see, among others, Loui, 1994, Karacapilidis and Gordon, 1996, Elhadad, 1995, McConachy and Zukerman, 1996) both because it is concerned with the effectiveness of arguments rather than their soundness and because, in the process of tailoring the arguments to the users, attention is paid to not purely logical attitudes, rather than just beliefs and knowledge. It is ....

Elhadad, M. (1995). Using argumentation in text generation. Journal of Pragmatics 24:189--220.

....of communicative goals, a NLG system must (a) select information from an underlying domain representation and determine how to structure and order that information (b) realize the content as sentences by selecting words and ordering them according to the syntactic constraints of the language. In [10], Elhadad presents a model of how argumentative intents can influence the decision making of a natural language generator at all levels of the generation process. Elhadad s model is based on abstract descriptive notions called topoi [3] gradual inference rules that capture commonsense relations ....

....language generator at all levels of the generation process. Elhadad s model is based on abstract descriptive notions called topoi [3] gradual inference rules that capture commonsense relations between evaluative judgments. For instance, in the real estate domain 11 one 11 All the examples in [10] are about a system advising students on whether to take or not certain courses. The examples are reformulated here in the real estate domain to facilitate comparison with examples in other topos could be: the less expensive a house is (other things being equal) the more people want to buy rent ....

[Article contains additional citation context not shown here]

Michael Elhadad. Using argumentation in text generation. Journal of Pragmatics, 24:189--220, 1995.

....they are cheaper. Although careful media allocation and coordination are basic steps in generating multimedia explanation, in this study, we have not yet addressed these issues. As far as media realization is concerned, for language, we are adapting techniques for generating arguments presented in (Elhadad 1995), whereas for graphics, we plan to use a tool, currently under development by our group, that designs graphics based on tasks that the users should be able to perform (e.g. search for certain objects, look up certain values) A final consideration. Since users, in interacting with an IDEA ....

Elhadad, M. 1995. Using argumentation in text generation.

.... intelligence community (Marcu 1996; Reiter et al. 1997) Also, many argumentative systems exist, although they are either focused on presenting valid arguments (Fox, Krause, ElvangGoransson 1993; Karacapidilis 1996; Vreeswijk 1997) or stress the generation of arguments in natural language (Elhadad 1995; Maybury 1993; Reed Long 1997) Zukerman, Korb, McConachy (1996) describe a system able to generate nice arguments tailored to the addressee, even though validity is again the evaluating measure used, an argument being nice when it can persuade though having some steps missing in its ....

Elhadad, M. 1995. Using Argumentation in Text Generation.

....difficult to abstract away from the syntactic constraints when negotiating to which syntactic slot a given conceptual attribute is to be mapped. mt2.tex; 10 06 1996; 17:35; no v. p. 4 Lexical Choice for Complex Noun Phrases 5 this pragmatic goal can affect the structure of the generated NP (cf. [15] for a discussion of the use of argumentation in generation in general. The LCNP therefore acts as a black box that receives as input a conceptual description for sets and individuals and is controlled by discursive and evaluative parameters to produce a syntactic tree for a complete NP. The ....

....of vague quantifiers, and justifies the addition of pragmatic annotations within the input to the LCNP. The particular model we have developed for the evaluative model is based on the theory of Argumentation Within Language (AWL) of Anscombre and Ducrot [1] and is described in detail in [15]. 4. Output of the LCNP The output of the LCNP is an input to the syntactic realization component surge. In this section, we describe the main features of the NP descriptions expected by surge. 4.1. Syntactic Structure of NPs Following [34] and [20] the syntactic description of NPs contains ....

M. Elhadad. Using argumentation in text generation. Journal of Pragmatics, 24:189--220, 1995.

No context found.

"Using Argumentation in Text Generation", Journal of Pragmatics 24, pp189220

No context found.

"Using Argumentation in Text Generation", Journal of Pragmatics 24, pp189220

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC