Modeling all dialogue system participants to generate empathetic responses

Haimowitz, Ira J.

doi:10.1016/0169-2607(91)90010-q

Cited by 6 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…such language increases the physician's (and therefore also the prospective system's) chances of having the patient hear the diagnosis and treatment recommendations' [20]. In her paper about the generation of 'empathetic responses' in a dialogue about medical treatment, Haimovitz [21] argues that to produce texts suited to the direct and the indirect users' needs, sentences with a given information content should be generated so as to 'stress favourable information while downplaying or offsetting unfavourable information'. This effect can be obtained by exploiting knowledge about the indirect user's 'concerns and worries', in addition to domain knowledge about the possible 'impact' of information items.…”

Section: Related Researchmentioning

confidence: 99%

“…Examples of strengths are; 'numerical measures of the impact of an utterance on the Hearer' in [21] and 'quantitative fuzzy measures' (a combination of 'support' and 'plausibility' of a proposition) in [53]. If modelled this way, the plan enrichment process involves some scalar value of 'mounting evidence' or 'increasing conviction' and stops when the system considers that the Hearer has been reasonably informed or persuaded.…”

Section: Limitations Of Knowledge Representationmentioning

confidence: 99%

“…To introduce empathy elements in the text, we modify the generation of individual sentences by applying heuristics of the following type (after each rule, the example from Section 3 that the rule is meant to solve is expressed): These heuristics are similar to those employed by Haimowitz to produce 'emotionally empathetic responses' [21]. There is, in fact, a surprising similarity between our data and Haimovitz's examples of 'hypothetical dialogue', such as the use of 'only three days' to downplay the length of hospital stay, or 'unfortunately, the Bactrim does leave a rash'.…”

Section: Re6ising the Atns Librarymentioning

confidence: 99%

See 2 more Smart Citations

Refining instructional text generation after evaluation

Rosis

Grasso

Berry

1999

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

In this paper, we describe how user-adapted explanations about drug prescriptions can be generated from already existing data sources. We start by illustrating the two-step approach employed in the first version of the natural language generator and the limitations of generated texts, that we discovered through analytical and empirical evaluations. We claim that, although style refinement would be needed in these texts, particular care should be devoted to implementing some of the persuasion techniques that doctors employ in their explanations. This would require either thoroughly revising the text planning techniques employed or converting to a multistep generation architecture. We justify why we selected this second alternative and propose some heuristics to repair problems found in the first version of the generator. Some final considerations about the advantages of this approach and the possibility of generalizing it to other domains conclude the paper.

show abstract

Section: Related Researchmentioning

confidence: 99%

Section: Limitations Of Knowledge Representationmentioning

confidence: 99%

Section: Re6ising the Atns Librarymentioning

confidence: 99%

See 1 more Smart Citation

Refining instructional text generation after evaluation

Rosis

Grasso

Berry

1999

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

show abstract

“…Similarly, the system should only choose the strategy of giving examples if the user will find the examples illustrative, i.e., the user knows the example concepts. Finally, knowledge about the user is useful in deciding which lexical items and syntactic structures should be used to express a concept in English (e.g., Jameson and Wahlster, 1982;Reithinger, 1987;Bateman and Paris, 1989;Haimowitz, 1990). Therefore, having information about the current user is important in providing meaningful explanations.…”

Section: The User Modelmentioning

confidence: 99%

Exploiting user feedback to compensate for the unreliability of user models

Moore

Paris²

1992

User Model User-Adap Inter

View full text Add to dashboard Cite

Abstract. Natural Language is apowerful medium for interacting with users, and sophisticated computer systems using natural language are becoming more prevalent. Just as human speakers show an essential, inbuilt responsiveness to their hearers, computer systems must "tailor" their utterances to users. Recognizing this, researchers devised user models and strategies for exploiting them in order to enable systems to produce the "best" answer for a particular user. and her bachelor's degree from the University of California in Berkeley. Her research interests include natural language generation and user modeling, discourse, expert system explanation, human-computer interaction, intelligent tutoring systems, machine learning, and knowledge acquisition. At Columbia University, she developed a natural language generation system capable of producing multi-sentential texts tailored to the users" level of expertise about the domain. At ISI, she has been involved in designing a flexible explanation facility that supports dialogue for an expert system shell. Because these efforts were largely devoted to investigating how a user model could be exploited to produce better responses, systems employing them typically assumed that a detailed and correct model of the user was available a priori, and that the information needed to generate appropriate responses was included in that model. However, in practice, the completeness and accuracy of a user model cannot be guaranteed. Thus, unless systems can compensate for incorrect or incomplete user models, the impracticality of building user models will prevent much of the work on tailoring from being successfully applied in real systems. In this paper, we argue that one way for a system to compensate for an unreliable user model is to be able to react to feedback from users about the suitability of the texts it produces. We also discuss how such a capability can actually alleviate some of the burden now placed on user modeling. F~inally, we present a text generation system that employs whatever information is available in its user model in an attempt to produce satisfactory texts, but is also capable of responding to the user's follow-up questions about the texts it produces.

show abstract

Affective Natural Language Generation

Rosis

Grasso

2000

Affective Interactions

View full text Add to dashboard Cite

Modeling all dialogue system participants to generate empathetic responses

Cited by 6 publications

References 7 publications

Refining instructional text generation after evaluation

Refining instructional text generation after evaluation

Exploiting user feedback to compensate for the unreliability of user models

Affective Natural Language Generation

Contact Info

Product

Resources

About