Mellish, C. O'Donnell, M., Oberlander, J. and Knott, A. An architecture for opportunistic text generation. In Proceedings of the 9th International Workshop on Natural Language Generation, INLG'98, Niagara-on-the-Lake, 1998, 28-37.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....techniques that lead to new goals. Although potentially less flexible, the recursive approach enables the use of efficient techniques (schemas, sequential execution) to compute the main text reliably and quickly. Another approach to the content adaptivity problem was explored in the ILEX system [Mellish et al. 1998] , which uses opportunistic planning to tailor the hypertext descriptions of museum exhibits. The planning mechanism is based on a structure called text potential a graph of facts connected with thematic and rhetorical relations. Facts are chosen depending on their connection to the focus of ....

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. An architecture for opportunistic text generation. In Proceedings of the International Natural Language Generation Workshop IWNLG'98, 1998.

....starts from the premise that discourse meaning is more than the sum of its parts (i.e. its constituent sentences or clauses) The question is how to get there. Work in the tradition of Rhetorical Structure Theory (RST) Mann Thompson, 1988) both in interpretation (Marcu, 2000) and generation (Mellish et al. 1998) views the additional meaning solely in terms of discourse relations that hold between adjacent text spans, treating discourse connectives as signalling types of discourse relations. How the basic text spans are assigned an interpretation, and how that interpretation might contribute to ....

Mellish, Chris, Mick O'Donnell, Jon Oberlander & Alistair Knott (1998). An Architecture for Opportunistic Text Generation. In Proc. of the 9th Intl. Workshop on NLG, pp. 28--37. Ontario, CA.

....aims at capturing the extra linguistic context of the user, including its most dynamic aspects, i.e. her minute to minute evolving intentions. The mechanism of competition for attention constitues an innovative way of treating the user s context. It can be compared with opportunistic NLG in [18], where system goals compete with user goals in order to generate at each moment the most effective museum object description. The set up of a virtual (agent) society allows us to scale this competition to large numbers of competing interests and to introduce the notion of community ware: virtual ....

Mellish, C., O'Donnell, M., Oberlander, J., Knott, A.: An Architecture for Opportunistic Text Generation. In Proceedings of the 9th Int. Workshop on Natural Language Generation, Niagara-on-the-Lake, Canada (1998) 28--37

....and Paris, 1995) inter alia. There have also been reactions to the standard planning architectures: Haller et al. Haller, 1994) Haller and Shapiro, 1996) for example, suggest a more reactive approach, echoed by criticisms of the top down approach to generation in general (Marcu, 1997a) (Mellish et al. 1998). Others such as (Meteer, 1991) Inui et al. 1992) de Rosis et al. 1997) and (Robin, 1994) have proposed a revision based approach to generation whereby a complete draft plan is created and then subsequently improved (this is similar to the select and repair method of the HealthDoc ....

, Niagra

....and Paris, 1995) inter alia. There have also been reactions to the standard planning architectures: Haller et al. Haller, 1994) Haller and Shapiro, 1996) for example, suggest a more reactive approach, echoed by criticisms of the top down approach to generation in general (Marcu, 1997a) (Mellish et al. 1998). Others such as (Meteer, 1991) Inui et al. 1992) de Rosis et al. 1997) and (Robin, 1994) have proposed a revision based approach to generation whereby a complete draft plan is created and then subsequently improved (this is similar to the select and repair method of the HealthDoc ....

, Niagra

....1. Introduction: Text Planning 1.1 The Task This paper presents some initial experiments using stochastic search methods for aspects of text planning. The work was motivated by the needs of the ILEX system for generating descriptions of museum artefacts (in particular, 20th Century jewellery) Mellish et al. 98] We present results on examples semi automatiCally generated from datastructures that exist within ILEX. Forming .a set. of facts about a piece of jewellery .into a structure that yields a coherent text is a non trivial problem. Rhetorical Structure Theory [Mann and Thompson 87] claims that a ....

....is an element of Hovy s work but is more apparent in. the planning work of Moore and Paris [Moore and Paris 93] This second approach will work well if there are strong goals in the domain which really can influence textual decisions. This is not always the case. For instance, in our ILEX domain [Mellish et al. 98] the system s goal is something very general like. say interesting things about item X subject to length and coherence constraints . The third approach, most obviously exemplified by [Marcu 97] is to Use some for m of explicit. search through possible trees, guided by heuristics about tree ....

Mellish, C., O'Donnell, M., Oberlander, J. and Knott, A., "An Architecture for Oppor- tunistic Text Generation", Proceedings of INLGW-98, 1998.

....in the archaic period, helping the visitor build a more coherent view of the collection. Following ILEX, the user model also contains scores indicating the educational value of each piece of information, as well as how likely it is for users of a particular type to find the information interesting [Mellish et al. 1998a; Oberlander et al. 1998] For each exhibit, the database typically contains more information than can be expressed in a description of reasonable length. The system attempts to convey only facts that have not been expressed in the past, and among those, it focuses on facts of high interest and ....

....two newlyweds. and friends is a canned string stored as the value of exhibit depicts ; the rest of the text is generated dynamically. Following ILEX, M PIRO associates with each fact in the database an interest, an importance, and an assimilation score (not shown in Figure 5) per user type [Mellish et al. 1998a; Oberlander et al. 1998] The interest score shows how likely it is for a visitor of a particular type to find the fact interesting. Domain experts, for example, may be interested to see references to published articles that discuss the selected exhibit, while casual visitors would probably find ....

[Article contains additional citation context not shown here]

C. Mellish, M. O'Donnell, J. Oberlander and A. Knott. "An Architecture for Opportunistic Text Generation". In Proceedings of the 9 th International Workshop on Natural Language Generation, Niagara-on-the-Lake, Ontario, Canada, 1998.

....sets and sequences if that was needed for a given class of applications. 3. 3 Example Instantiations and Surface Syntaxes The following instantiation of the above primitive types is used in a Prolog interface to a simpli ed ILEX knowledge base about pieces of modern jewellery at Edinburgh [15]. SemRole = farg1; arg2; pred; nuc; sat; reln; inv arg1; inv arg2; inv nuc; inv sat; nameg SemP red = fentity; fact; reln; jewellery; spatio temporal thing : g In fact, what is being modelled here is what the ILEX project calls the content potential a language oriented repository of the ....

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. An architecture for opportunistic text generation. In Proceedings of the 9th International Workshop on Natural Language Generation, pages 28-37, Niagra-on-the-Lake, 1998.

....the question of integrating referring and informing, although rather briefly, and without much detail. This paper will extend upon his discussion, and describe its role in ILEX, a text generation system which delivers descriptions of entities on line from an underlying knowledgebase (see Mellish et al. 1998). ILEX is at present generating descriptions in the museum domain, in particular, that of 20th Century jewellery. Our focus on this topic has grown out of the need to integrate two strands of research within ILEX. One strand involves the work on anaphora by Janet Hitzeman. She implemented a module ....

Mellish, C., O'Donnell, M., Oberlander, J. and Knott, A. 1998 "An architecture for opportunistic text generation". Proceedings of the 9th International Workshop on Natural Language Generation. 5-7 August 1998. Prince of Wales Hotel, Niagara-on-the-Lake, Ontario, Canada.

....has already assimilated can be taken into account in the current description. For instance, the description of the object currently being viewed can make use of comparisons and contrasts to previously viewed objects, while omitting any background information that the visitor has already been told [6, 7, 8]. Dynamic hypertext makes it possible for the generation system to pursue its own agenda of educational and communicative goals, while allowing the user the freedom to browse the collection of objects in any order, as in a normal hypermedia system. The aim is to reproduce the kind of descriptions ....

....reproduce the kind of descriptions that a real curator might give, were the visitor to have one at their elbow. Opportunistic text tailoring is achieved in ILEX via the use of referring expressions, comparison expressions, nominal anaphora and approaches derived from rhetorical structure theory [4, 6, 7] 1 ) The aim of the evaluation was to attempt to assess the effect of intelligent label generation upon several types of learning outcome. Dynamic and static versions of the intelligent labelling explorer (ILEX) system were compared. The goal was to attempt to pin down or isolate to some ....

Mellish, C., O'Donnell, M., Oberlander, J. and Knott, A. (1998b) An architecture for opportunistic text generation. In Proceedings of the 9th International Generation Workshop, Niagra-on-the-Lake, Ontario, Canada.

No context found.

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. 1998b. An architecture for opportunistic text generation. In Proceedings of the 9th International Workshop on Natural Language Generation, Ontario, Canada.

....arcs indicate resumption relations: links from an entity chain to the sentence which introduces it. Note that these arcs do not have to link adjacent entity chains, and can cross one another. The model of text structure just outlined has been implemented in the ILEX 2 text generation system; see Mellish et al. (1998) for details. An example of a text generated by the 3 In this respect, our proposal is similar to that made by Mooney et al. (1990) 4 A working de nition of what it is for a fact to be about a certain entity is given in Mellish et al. (1998) 8 system is given below: 6) 1) This piece is a ....

....implemented in the ILEX 2 text generation system; see Mellish et al. (1998) for details. An example of a text generated by the 3 In this respect, our proposal is similar to that made by Mooney et al. (1990) 4 A working de nition of what it is for a fact to be about a certain entity is given in Mellish et al. (1998). 8 system is given below: 6) 1) This piece is a necklace. 2) It was designed by a jeweller called Jessie King. 3) It was designed in 1905. 4) It is made of silver and enamel. 5) Jessie King was a famous designer. 6) She was Scottish, 7) but she worked in London. 8) It was in London ....

Mellish, C., O'Donnell, M., Oberlander, J., and Knott, A. (1998). An architecture for opportunistic text generation. In Proceedings of the ninth International Workshop on Natural Language Generation, pages 28-37, Montreal.

No context found.

Mellish, C. O'Donnell, M., Oberlander, J. and Knott, A. An architecture for opportunistic text generation. In Proceedings of the 9th International Workshop on Natural Language Generation, INLG'98, Niagara-on-the-Lake, 1998, 28-37.

No context found.

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. 1998. An architecture for opportunistic text generation. In Proceedings of the Ninth International Workshop on Natural Language Generation, Niagara-on-the-Lake, Ontario, Canada.

No context found.

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. 1998. An architecture for opportunistic text generation. In Proceedings of the Ninth International Workshop on Natural Language Generation, Niagara-on-the-Lake, Ontario, Canada.

No context found.

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. 1998. An architecture for opportunistic text generation. In Proceedings of the Ninth International Workshop on Natural Language Generation, Niagara-on-the-Lake, Ontario, Canada.

No context found.

Chris Mellish, Mick O'Donnell, Jon Oberlander, and Alistair Knott. 1998. An architecture for opportunistic text generation. In Proc. of the 9th International Workshop on Natural Language Generation.

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