Wannet, L., E. Hovy. 1996. The HealthDoc Sentence Planner. In Proc. of the 8th Interna- tional Workshop on Natural Language Generation, Herstmonceux Castle, 1-10. 137  Home/Search   Document Details and Download   Summary   Related Articles   Check

....the previous two [25, 35] In this intermediate phase, local refinement techniques are applied to increase the quality of the generated text. For example: determining the sentence scope, aggregating clauses or elements internal to a clause, defining how to realize rhetorical relations and so on [40, 13, 50, 42]. As most systems that have been developed are still in a prototype phase, not many of them have been evaluated in a real world environment, corresponding author. 1 to assess whether the texts produced correspond to what speakers and addressees would expect from them. Moreover, existing ....

....described in the previous Section, by applying them in turn to the first version of OPADE s output. To come to an integrated new system, we have to define how to combine the described techniques and which is the appropriate architecture to combine the various modules: a pipeline [42] a blackboard [50], a parallel processing [40] or something else. Finally, we should evaluate if the new texts are really preferred by Indirect Users: evaluation experiments often reserve surprises; an enlighting study about whether naturalness increases effectiveness of automated telephone dialogues, for ....

L. Wanner and E. Hovy. The HealthDoc Sentence Planner. In Proceedings of the 8th International Workshop on Natural Language Generation, pages 1--10, 1996.

....information from the old. Each operator is independent of the others and several can be applied in succession to the input templates. Each of the seven major operators is further subdivided to cover various modi cations to its input. This procedure is similar to the sentence planner of [Wanner and Hovy, 1996]. A summary operator encodes a means for linking information in two di erent templates. Often it results in the synthesis of new information. For example, a generalization may be formed from two independent facts. Alternatively, since we are summarizing reports written over time, highlighting how ....

....from a so called master document and generates a concise summary of the master document s full document content. To do so, HealthDoc s sentence planner also needs to perform LRR in order to remove infelicities of phrasing and lexicalization due to segment combination. The sentence planner [Wanner and Hovy, 1996] contains a list of rules that operate on the internal representations, transforming them in order to ensure better output, in exactly the same way as the LRR operators do in Chapter 5. 141 Chapter 11 Applications and future work This chapters contains three sections. The rst two deal with ....

Leo Wanner and Eduard Hovy. The HealthDoc sentence planner. In Proceedings of the 8th International Workshop on Natural Language Generation, pages 1-10, Herstmonceux, England, June 1996.

....information from the old. Each operator is independent of the others and several can be applied in succession to the input templates. Each of the seven major operators is further subdivided to cover various modifications to its input. This procedure is similar to the sentence planner of [Wanner and Hovy, 1996]. A summary operator encodes a means for linking information in two di#erent templates. Often it results in the synthesis of new information. For example, a generalization may be formed from two independent facts. Alternatively, since we are summarizing reports written over time, highlighting how ....

....from a so called master document and generates a concise summary of the master document s full document content. To do so, HealthDoc s sentence planner also needs to perform LRR in order to remove infelicities of phrasing and lexicalization due to segment combination. The sentence planner [Wanner and Hovy, 1996] contains a list of rules that operate on the internal representations, transforming them in order to ensure better output, in exactly the same way as the LRR operators do in Chapter 5. 141 Chapter 11 Applications and future work This chapters contains three sections. The first two deal with ....

Leo Wanner and Eduard Hovy. The HealthDoc sentence planner. In Proceedings of the 8th International Workshop on Natural Language Generation, pages 1--10, Herstmonceux, England, June 1996.

....is gained. We do not elaborate the advantages further here; see, for example, Panaget 1994] What are the specific decisions to be made by the sentence planner We think it is important to separate the formative decisions from the motivations that lead to the particular choices. Following Wanner and Hovy [1996], a sentence planner has to make the following decisions: Fine grained discourse structuring, including discourse marker choice; sentence grouping and sentence content determination; clause internal structuring; choice of referring expressions; lexical choice. 1 Two groups of considerations are ....

....order of making decisions in sentence planning will impose limitations on the expressiveness of the system. Accordingly, we advocate a flexible order of decision making, as it can be realized in a blackboard based architecture such as proposed by DIOGENES [Nirenburg et al. 1989] and HealthDoc [Wanner and Hovy 1996]. Moreover, the individual modules or knowledge sources should rely on declarative representations as much as possible; otherwise the control process becomes extremely complicated. And one of the declarative sources of information, we feel, should be a lexicon that assembles specifically the ....

[Article contains additional citation context not shown here]

Wanner, L., E. Hovy. 1996. The HealthDoc Sentence Planner. In Proc. of the 8th International Workshop on Natural Language Generation, Herstmonceux Castle, 1-10.

....order, etc. Paradigmatic features: Features that label the differences between similar markers sharing the same applicability conditions, such as stylistic features and degrees of emphasis. Very briefly, we see discourse marker choice as one aspect of the sentence planning task (e.g. [Wanner, Hovy 1996]) In order to account for the intricate interactions between marker choice and other generation decisions, the idea is to employ DiMLex as a declarative resource supporting the sentence planning process, which comprises determining sentence boundaries and sentence structure, linear ordering of ....

L. Wanner, E. Hovy. "The HealthDoc sentence planner. " In: Proceedings of the Eighth International Workshop on Natural Language Generation, Herstmonceux Castle, June 1996.

....generation is not performed in any real sense, all referring expressions being always generic or proper . There is no use of centering or salience. CD SP SR Job description builder ffl ffl Job description generator ffl 3. 6 HealthDoc HealthDoc ( DHW95] DHH97] HDHP97] HW96] Par97] WH96] Wil95] is an unusual system in that it does not do generation of language from another mode, but rather selects and repairs existing text parts. The input to HealthDoc is a full natural language text, which (allegedly) contains every piece of information which could be wanted, and the output ....

L. Wanner and E. Hovy. The HealthDoc Sentence Planner. In INLG'96 [INL96], pages 1--10.

....as MIGRAINE (Buchanan et al. 1995) and Piglet (Binsted, Cawsey, and Jones, 1995) their applications have not required that they combine several independent but inter related text plans into a single integrated message. The work most closely related to ours is HealthDoc (Hirst and DiMarco, 1996; Wanner and Hovy, 1996); however, HealthDoc (currently under development) focuses on editing sentences selected from a master text, such as by inserting pronouns or by deleting references to propositions that do not appear earlier in the selected text. Constructing Effective Message Units The nature of trauma ....

L. Wanner and E. Hovy (1996). The HealthDoc sentence planner. In Proceedings of the International Workshop on Natural Language Generation, 1--10.

....the previous two [25, 35] In this intermediate phase, local refinement techniques are applied to increase the quality of the generated text. For example: determining the sentence scope, aggregating clauses or elements internal to a clause, defining how to realize rhetorical relations and so on [40, 13, 50, 42]. As most systems that have been developed are still in a prototype phase, not many of them have been evaluated in a real world environment, to assess whether the texts produced correspond to what speakers and addressees would expect from them. Moreover, existing evaluations often consist of an ....

....described in the previous Section, by applying them in turn to the first version of OPADE s output. To come to an integrated new system, we have to define how to combine the described techniques and which is the appropriate architecture to combine the various modules: a pipeline [42] a blackboard [50], a parallel processing [40] or something else. Finally, we should evaluate if the new texts are really preferred by Indirect Users: evaluation experiments often reserve surprises; an enlighting study about whether naturalness increases effectiveness of automated telephone dialogues, for ....

L. Wanner and E. Hovy. The HealthDoc Sentence Planner. In Proceedings of the 8th International Workshop on Natural Language Generation, pages 1--10, 1996.

....in strategic component [Mann and Moore, 1980, Dale, 1992, Horacek, 1992] Logical derivations were used to combine clauses and remove easily inferable clauses in [Mann and Moore, 1980] In such systems, aggregation decisions are made without lexical information. Newer systems, such as [Shaw, 1995, Wanner and Hovy, 1996, Huang and Fiedler, 1997] use a sentence planner to make decisions at clause level between the strategic and tactical component. With the exception of [Scott and de Souza, 1990] and [Robin, 1995] most research in aggregation did not transform clauses into modifiers, such as adjectives, PP, or ....

Wanner, L. and Hovy, E. The HealthDoc sentence planner. In Proc. of the 8th International Natural Language Generation Workshop, pages 1--10, Sussex, UK.

....architecture that enables such an approach and facilitates near linear time processing with realistic texts. 1 Introduction This paper addresses the area of text generation known as microplanning [Levelt1989, Panaget1994, Huang and Fiedler1996] or sentence planning [Rambow and Korelsky1992] [Wanner and Hovy1996]. Microplanning involves low level discourse structuring and marking, sentence boundary planning, clause internal structuring and all of the varied subtasks involved in lexical choice. Conventional wisdom dictates that these complex tasks be modularized and treated separately: Since sentence ....

....Since sentence planning tasks are not single step operations, since they do not have to be performed in strict sequence, and since the planner s operation is non deterministic, this suggests that each sentence planning task should be implemented by a separate module or by several modules. [Wanner and Hovy1996] Such an argument is natural if generation is viewed as a set of coarse grained tasks. Indeed, with the exception of a few researchers ( Elhadad et al..1997] and the incrementalists listed below) the taskoriented view is standard in the generation community. Unfortunately, task oriented generation ....

[Article contains additional citation context not shown here]

L. Wanner and E. Hovy. 1996. The healthdoc sentence planner. Proc. Eighth International Natural Language Generation Workshop (INLG-96).

....the traditional jargon and refer to the task of content organization as text planning . Moore and Paris, 1993, Meteer, 1992, Moore, 1995 ] and derive the set of rhetorical relations that hold between every pair of semantic units. For very rich KBs, such as that used in the HealthDoc Project [ Wanner and Hovy, 1996, Hovy and Wanner, 1996, DiMarco and Foster, 1997, DiMarco et al. 1997, Hirst et al. 1997 ] one can simply extract these relations directly, because they are explicitly represented. Each of the alternatives described above has been already discussed in the literature to a greater or lesser ....

....to the task of content organization as text planning . Moore and Paris, 1993, Meteer, 1992, Moore, 1995 ] and derive the set of rhetorical relations that hold between every pair of semantic units. For very rich KBs, such as that used in the HealthDoc Project [ Wanner and Hovy, 1996, Hovy and Wanner, 1996, DiMarco and Foster, 1997, DiMarco et al. 1997, Hirst et al. 1997 ] one can simply extract these relations directly, because they are explicitly represented. Each of the alternatives described above has been already discussed in the literature to a greater or lesser extent. Therefore, for the ....

[Article contains additional citation context not shown here]

Leo Wanner and Eduard Hovy. The HealthDoc sentence planner. In Proceedings of the Eighth International Natural Language Generation Workshop (INLG--96), pages 1--10, Herstmonceux, UK, June 12--15 1996.

....necessary repairs, with each type of repair performed by an independent repair module. The sentence planner is based on a blackboard architecture in which individual repair modules communicate and resolve their conflicts with one another. The architecture is described in greater detail by Hovy and Wanner (1996) and Wanner and Hovy (1996) Four repair modules are being built in the first phase of the main HealthDoc project: for discourse structuring, aggregation to remove redundancies, reference restoration using pronouns, and constituent re ordering. WebbeDoc: An application of the HealthDoc approach ....

....with each type of repair performed by an independent repair module. The sentence planner is based on a blackboard architecture in which individual repair modules communicate and resolve their conflicts with one another. The architecture is described in greater detail by Hovy and Wanner (1996) and Wanner and Hovy (1996). Four repair modules are being built in the first phase of the main HealthDoc project: for discourse structuring, aggregation to remove redundancies, reference restoration using pronouns, and constituent re ordering. WebbeDoc: An application of the HealthDoc approach to the automated ....

Wanner, Leo and Hovy, Eduard (1996). "The HealthDoc sentence planner." Proceedings of the Eighth International Workshop on Natural Language Generation, Brighton, UK, June 1996.

....the generation task. We propose a new modularity based on natural linguistic phenomena and overview how it is implemented in the Mikrokosmos text generator. 1 Introduction This paper addresses the area of text generation that has come to be known as sentence planning (Rambow and Korelsky, 1992) (Wanner and Hovy, 1996). Sentence planning involves discourse structuring and marking, sentence boundary planning, clause internal structuring and all of the varied subtasks involved in lexical choice. Conventional wisdom dictates that these complex tasks be modularized and treated separately: Since the sentence ....

....sentence planning tasks listed above are not single step operations, since they do not have to be performed in strict sequence, and since the planner s operation is non deterministic, this suggests that each sentence planning task should be implemented by a separate module or by several modules. (Wanner and Hovy, 1996) In contrast, we argue that such a functional division sets up artificial barriers, with the result that fighting those barriers has become a pervasive topic in the generation literature, a topic that becomes moot when our approach is followed. The very words of the argument quoted above not ....

[Article contains additional citation context not shown here]

L. Wanner and E. Hovy. 1996. The healthdoc sentence planner. Proc. Eighth International Natural Language Generation Workshop (INLG-96).

....necessary repairs, with each type of repair performed by an independent repair module. The sentence planner is based on a blackboard architecture in which individual repair modules communicate and resolve their conflicts with one another. The architecture is described in greater detail by Hovy and Wanner (1996) andWanner and Hovy (1996) Four repair modules are being built in the first phase of the main HealthDoc project: for discourse structuring and rhetorical relations, aggregation to remove redundancies, reference restoration using pronouns, and constituent re ordering. The first phase: Text ....

....with each type of repair performed by an independent repair module. The sentence planner is based on a blackboard architecture in which individual repair modules communicate and resolve their conflicts with one another. The architecture is described in greater detail by Hovy and Wanner (1996) andWanner and Hovy (1996). Four repair modules are being built in the first phase of the main HealthDoc project: for discourse structuring and rhetorical relations, aggregation to remove redundancies, reference restoration using pronouns, and constituent re ordering. The first phase: Text generation by selection only in ....

Wanner, Leo and Hovy, Eduard (1996). "The HealthDoc sentence planner." Proceedings of the Eighth International Workshop on Natural Language Generation, Brighton, UK, June 1996.

No context found.

Wannet, L., E. Hovy. 1996. The HealthDoc Sentence Planner. In Proc. of the 8th Interna- tional Workshop on Natural Language Generation, Herstmonceux Castle, 1-10. 137

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