A Probabilistic Approach to the Interpretation of Spoken Utterances

Zukerman, Ingrid; Makalic, Enes; Niemann, Michael; George, Sarah

doi:10.1007/978-3-540-89197-0_53

Cited by 8 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm generates interpretations until time runs out (in our case, until a certain number of iterations has been executed). In Steps 1-5, Algorithm 1 processes each sentence separately according to the interpretation process for single sentences described in (Zukerman et al, 2008). 1 Charniak's probabilistic parser 2 is applied to generate parse trees for each sentence in the sequence.…”

Section: Interpreting a Sequence Of Utterancesmentioning

confidence: 99%

“…One sentence. The probability of a UCG generated from a sentence T is estimated as described in (Zukerman et al, 2008), resulting in…”

Section: Estimating the Probability Of A Merged Interpretationmentioning

confidence: 99%

“…This observation, which was validated in our corpus study (Section 4), motivates the mechanism for the interpretation of a sequence of sentences presented in this paper. Our mechanism extends our probabilistic process for interpreting single spoken utterances (Zukerman et al, 2008) in that (1) it determines which sentences in a sequence are related, and if so, combines them into an integrated interpretation; and (2) it provides a formulation for estimating the probability of an interpretation of a sentence sequence, which supports the selection of the most probable interpretation. Our evaluation demonstrates that our mechanism performs well in understanding textual sentence pairs of different length and level of complexity, and highlights particular aspects of our algorithms that require further improvements (Section 4).…”

Section: Introductionmentioning

confidence: 99%

“…We focus on imperative UCGs because they contain the actions that the robot is required to perform; these actions incorporate relevant information from declarative UCGs. ICGs are generated by nominating different instantiated concepts and relations from the system's knowledge base as potential realizations for each concept and relation in a UCG (Zukerman et al, 2008); each UCG can generate many ICGs. Since this paper focuses on the generation of UCG sequences, the generation of ICGs will not be discussed further.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Towards the interpretation of utterance sequences in a dialogue system

Zukerman

Gupta

et al. 2009

Proceedings of the SIGDIAL 2009 Conference on the 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue -

Self Cite

View full text Add to dashboard Cite

This paper describes a probabilistic mechanism for the interpretation of sentence sequences developed for a spoken dialogue system mounted on a robotic agent. The mechanism receives as input a sequence of sentences, and produces an interpretation which integrates the interpretations of individual sentences. For our evaluation, we collected a corpus of hypothetical requests to a robot. Our mechanism exhibits good performance for sentence pairs, but requires further improvements for sentence sequences.

show abstract

Section: Interpreting a Sequence Of Utterancesmentioning

confidence: 99%

“…One sentence. The probability of a UCG generated from a sentence T is estimated as described in (Zukerman et al, 2008), resulting in…”

Section: Estimating the Probability Of A Merged Interpretationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Towards the interpretation of utterance sequences in a dialogue system

Zukerman

Gupta

et al. 2009

Proceedings of the SIGDIAL 2009 Conference on the 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue -

Self Cite

View full text Add to dashboard Cite

show abstract

“…We note that our approach does not apply any domain knowledge as e.g. that by (Zukerman et al, 2008). We see approach as a shallow method in which the domain knowledge is encoded implicitly in the graph of coherence relations.…”

Section: Preference-based Presentationmentioning

confidence: 99%

What Readers want to Experience - An Approach to Quantify Conversational Maxims with Preferences for Reading Behaviour

Knäusl¹,

Ludwig²

2013

Proceedings of the 5th International Conference on Agents and Artificial Intelligence

View full text Add to dashboard Cite

Abstract:Searching information on web pages is a tedious task for users as web servers provide complete web pages and do not tailor their content to the user's current information need. This leaves an enormous amount of workload for the user and influences his emotional attitude towards the whole task even if a search engine has filtered pages that are relevant to a user query. In this paper, we propose an approach to adapt the response to queries to user preferences for his reading experience in order to leverage the problem of information overload. With these preferences, it is possible to select the most preferred content from a web page. In our view, the preferences are a quantitative way to express conversational maxims. We present our experimental approach to learn these preferences from annotated browsing sessions and introduce a decision strategy for the selection of content on the basis of the learned preferences.

show abstract

Towards Interpreting Task-Oriented Utterance Sequences

Zukerman

2009

AI 2009: Advances in Artificial Intelligence

View full text Add to dashboard Cite

A Probabilistic Approach to the Interpretation of Spoken Utterances

Cited by 8 publications

References 9 publications

Towards the interpretation of utterance sequences in a dialogue system

Towards the interpretation of utterance sequences in a dialogue system

What Readers want to Experience - An Approach to Quantify Conversational Maxims with Preferences for Reading Behaviour

Towards Interpreting Task-Oriented Utterance Sequences

Contact Info

Product

Resources

About