A task-based evaluation of the TRAINS-95 dialogue system

Sikorski, Teresa; Allen, James F.

doi:10.1007/3-540-63175-5_48

Cited by 14 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TRAINS project based on the human-human Natural Spoken Dialogue based on task-oriented spoken dialogues. TRAINS corpus [26] consists of six and a half hours of speech, about 5900 speaker turns and 55000 transcribed words. On the other hand, human-machine conversation using LPJudges and LPBots get a lower score as compared to AINI and IM human users.…”

Section: Resultsmentioning

confidence: 99%

Using Gunnnig-Fog Index to Assess Instant Messages Readability from ECAs

Goh

Fung

Depickere

et al. 2007

Third International Conference on Natural Computation (ICNC 2007) Vol V

View full text Add to dashboard Cite

show abstract

Section: Resultsmentioning

confidence: 99%

Using Gunnnig-Fog Index to Assess Instant Messages Readability from ECAs

Goh

Fung

Depickere

et al. 2007

Third International Conference on Natural Computation (ICNC 2007) Vol V

View full text Add to dashboard Cite

show abstract

“…Like the Map Task corpus, the TRAINS corpus has also been categorized based on speech acts common to the discourse (Sikorski & Allen 1997). These speech acts are generally related to the corpus' purpose of building transportation plans for later use.…”

Section: Dimension 4: Directional Vs Non-directional Discoursementioning

confidence: 99%

Multi-dimensional register classification using bigrams

Crossley

Louwerse

2007

IJCL

View full text Add to dashboard Cite

A corpus linguistic analysis investigated register classification using frequency of bigrams in nine spoken and two written corpora. Four dimensions emerged from a factor analysis using bigram frequencies shared across corpora: (1) Scripted vs. Unscripted Discourse, (2) Deliberate vs. Unplanned Discourse, (3) Spatial vs. Non-Spatial Discourse, and (4) Directional vs. Non-Directional Discourse. These findings were replicated in a second analysis. Both analyses demonstrate the strength of bigrams for classifying spoken and written registers, especially in locating distinct collocations among spoken corpora, as well as revealing syntactic and discourse features through a data-driven approach.

show abstract

“…In the eighties and nineties of the previous century most of these relied on speech or keyboard input and produced either speech or text output only. The TRAINS dialogue system can be considered a classic example [28]. Since the start of this century the study and development of embodied conversational agents and humanoid robotics has lead to consider also nonverbal means of communication in tandem with speech and natural language [6] (see also the proceedings of the Intelligent Virtual Agents conference).…”

Section: Introductionmentioning

confidence: 99%

Multimodal end-of-turn prediction in multi-party meetings

Kok

Heylen

2009

Proceedings of the 2009 International Conference on Multimodal Interfaces

View full text Add to dashboard Cite

One of many skills required to engage properly in a conversation is to know the appropiate use of the rules of engagement. In order to engage properly in a conversation, a virtual human or robot should, for instance, be able to know when it is being addressed or when the speaker is about to hand over the turn. The paper presents a multimodal approach to end-of-speaker-turn prediction using sequential probabilistic models (Conditional Random Fields) to learn a model from observations of real-life multi-party meetings. Although the results are not as good as expected, we provide insight into which modalities are important when taking a multimodal approach to the problem based on literature and our own results.

show abstract

A task-based evaluation of the TRAINS-95 dialogue system

Cited by 14 publications

References 4 publications

Using Gunnnig-Fog Index to Assess Instant Messages Readability from ECAs

Using Gunnnig-Fog Index to Assess Instant Messages Readability from ECAs

Multi-dimensional register classification using bigrams

Multimodal end-of-turn prediction in multi-party meetings

Contact Info

Product

Resources

About