Dialogue Act Detection Using Contextual Knowledge

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F18%3A00322128" target="_blank" >RIV/68407700:21230/18:00322128 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/68407700:21730/18:00322128

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Dialogue Act Detection Using Contextual Knowledge

Popis výsledku v původním jazyce

Dialogue act detection is an important part of the conversational systems. It describes a user's intention which is one of the features the system uses to select the next action. In the comparison to the intent detection, dialogue acts are more general, and they do not depend on the current topic. We compare several model architectures which use a single utterance as an input, multiple concatenated utterances or utterance window in which the single utterances are processed sequentially. The experiments are conducted on a Switchboard dataset which consists of transcribed telephone conversations. The resulted system is used as part of the Alquist bot which is one of the competitors of the Alexa Prize 2018.

Název v anglickém jazyce

Dialogue Act Detection Using Contextual Knowledge

Popis výsledku anglicky

Dialogue act detection is an important part of the conversational systems. It describes a user's intention which is one of the features the system uses to select the next action. In the comparison to the intent detection, dialogue acts are more general, and they do not depend on the current topic. We compare several model architectures which use a single utterance as an input, multiple concatenated utterances or utterance window in which the single utterances are processed sequentially. The experiments are conducted on a Switchboard dataset which consists of transcribed telephone conversations. The resulted system is used as part of the Alquist bot which is one of the competitors of the Alexa Prize 2018.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the International Student Scientific Conference Poster – 22/2018

ISBN

978-80-01-06428-3

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

—

Název nakladatele

Czech Technical University in Prague

Místo vydání

Praha

Místo konání akce

Praha

Datum konání akce

10. 5. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—