On the Means for Clarification in Dialogue

Purver, Matthew; Ginzburg, Jonathan; Healey, Patrick G. T.

doi:10.1007/978-94-010-0019-2_11

Cited by 44 publications

(62 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Empirical work such as Ginzburg (1998) and Purver et al (2002) indicates that questioners do not usually refer back to questions which are very distant, and this was consistent with our data. In particular Purver et al analyzed the English dialogue transcripts of the British National Corpus, finding that clarification request source separation (CSS, the distance between a question and the question or answer which it is attempting to clarify) was at most 15 sentences and usually less than 10 sentences.…”

Section: Clarification Recognition Algorithmsupporting

confidence: 94%

Implementing clarification dialogues in open domain question answering

Boni

Manandhar

2005

Nat. Lang. Eng.

View full text Add to dashboard Cite

We examine the implementation of clarification dialogues, a mechanism for ensuring that question answering systems take into account user goals by allowing them to ask series of related questions either by refining or expanding on previous questions with follow-up questions, in the context of open domain Question Answering systems. We develop an algorithm for clarification dialogue recognition through the analysis of collected data on clarification dialogues and examine the importance of clarification dialogue recognition for question answering. The algorithm is evaluated and shown to successfully recognize the start and continuation of clarification dialogues in 94% of cases. We then show the usefulness of the algorithm by demonstrating how the recognition of clarification dialogues can simplify the task of answer retrieval.

show abstract

Section: Clarification Recognition Algorithmsupporting

confidence: 94%

Implementing clarification dialogues in open domain question answering

Boni

Manandhar

2005

Nat. Lang. Eng.

View full text Add to dashboard Cite

show abstract

“…Even these categories, however, can take a variety of surface forms. P2NTRIs (or clarification requests (CRs), see e.g., Ginzburg & Cooper, 2004) can appear not only as whwords as in (4), but short fragments (6), longer reprises or echoes (but not necessarily verbatim) (7), and more explicit or conventional indicators (8)- (9) (Purver, Ginzburg, & Healey, 2003): (6) Healey et al (2005) present a protocol for coding repair in interaction which identifies the different CA types of repair described above. Reliability of the protocol was shown to be encouraging-in an exercise re-coding a corpus of examples from the CA literature, 75% were assigned the same category as in the original-although detection agreement rates were not reported.…”

Section: Types Of Repairmentioning

confidence: 99%

“…Colman and Healey (2011) and McCabe et al (2013) report inter-annotator agreement of c. 75% kappa. BNC-PGH is annotated only for other-repair initiation P2NTRIs (Purver et al, 2003 report 75-95% kappa); MAPTASK similarly provides information on P2NTRIs (via check tags) but not self-repair. SWBD, BNC, and MAPTASK provide gold-standard part-of-speech (POS) tags; we tagged the PCC using the Stanford POS tagger (Toutanova, Klein, Manning, & Singer, 2003).…”

Section: Annotationmentioning

confidence: 99%

“…In order to detect NTRIs, we define a set of turn-level features that could be extracted from transcripts automatically and that encode either specific NTRI characteristics (e.g., presence of clarificational words like "pardon") or more general parallelism features between the turn to be classified and the previous turn by other and same speaker. (This assumes antecedents of clarification are in the immediately preceding turn; Purver et al [2003] found this to cover 85% of cases.) We then train a standard supervised discriminative classifier using these features to detect NTRI turns.…”

Section: Detecting Other-repairmentioning

confidence: 99%

See 1 more Smart Citation

Computational Models of Miscommunication Phenomena

Purver

Hough

Howes

2018

Topics in Cognitive Science

Self Cite

View full text Add to dashboard Cite

Miscommunication phenomena such as repair in dialogue are important indicators of the quality of communication. Automatic detection is therefore a key step toward tools that can characterize communication quality and thus help in applications from call center management to mental health monitoring. However, most existing computational linguistic approaches to these phenomena are unsuitable for general use in this way, and particularly for analyzing human-human dialogue: Although models of other-repair are common in human-computer dialogue systems, they tend to focus on specific phenomena (e.g., repair initiation by systems), missing the range of repair and repair initiation forms used by humans; and while self-repair models for speech recognition and understanding are advanced, they tend to focus on removal of "disfluent" material important for full understanding of the discourse contribution, and/or rely on domain-specific knowledge. We explain the requirements for more satisfactory models, including incrementality of processing and robustness to sparsity. We then describe models for self- and other-repair detection that meet these requirements (for the former, an adaptation of an existing repair model; for the latter, an adaptation of standard techniques) and investigate how they perform on datasets from a range of dialogue genres and domains, with promising results.

show abstract

“…Misunderstandings and non-understandings are much less frequent in human-human dialogue than in human-machine dialogue [11]. When they do occur, it is less often because a dialogue participant has misheard an utterance and more often that she has confusions about her conversational partner's underlying intent [12].…”

Section: A Aptness Of An Embedded Woz Corpusmentioning

confidence: 99%

Naturalistic Dialogue Management for Noisy Speech Recognition

Passonneau

Epstein

Ligorio

2012

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

With naturalistic dialogue management, a spoken dialogue system behaves as a human would under similar conditions. This paper reports on an experiment to develop naturalistic clarification strategies for noisy speech recognition in the context of spoken dialogue systems. We collected a wizard-of-Oz corpus in which human wizards with access to a rich set of clarification actions made clarification decisions online, based on human-readable versions of system data. The experiment compares an evaluation of calls to a baseline system in a library domain with calls to an enhanced version of the system. The new system has a clarification module based on the wizard data that is a decision tree constructed from three machine-learned models. It replicates the wizards' ability to ground partial understandings of noisy input and to build upon them. The enhanced system has a significantly higher rate of task completion, greater task success and improved efficiency.

show abstract

On the Means for Clarification in Dialogue

Cited by 44 publications

References 6 publications

Implementing clarification dialogues in open domain question answering

Implementing clarification dialogues in open domain question answering

Computational Models of Miscommunication Phenomena

Naturalistic Dialogue Management for Noisy Speech Recognition

Contact Info

Product

Resources

About