UPDATING MODELS OF CLASSIFIERS OF UNDERSTANDING LANGUAGE BASED ON CROWDSOURCING Russian patent published in 2019 - IPC G10L15/22 

Abstract RU 2699587 C2

FIELD: physics.

SUBSTANCE: group of inventions relates to language understanding classifiers models updating means. To this end, a method of updating models of language understanding classifiers, in which digital voice input is received from a user of a computing device through a microphone of a computing device. Natural language processing using voice input is used to detect user voice request. Upon establishing that the user voice request is inconsistent with one of the plurality of given voice commands in determining the personal digital assistant circuit, using the tag GUI to receive the user selection: at least one intention from a plurality of available intentions and/or at least one position for said at least one intent. Set of marked data is generated by pairing the user's voice request and user selection and is used to update the language understanding classifier.

EFFECT: technical result is higher accuracy of user commands recognition.

15 cl, 12 dwg

Similar patents RU2699587C2

Title Year Author Number
ELECTRONIC DEVICE AND ITS CONTROL METHOD 2020
  • Lee, Eunji
  • Ko, Hyeonmok
  • Lee, Kyenghun
  • Jang, Saebom
  • Jung, Pureum
  • Choi, Sungja
  • Paeon, Changho
  • Hong, Jiyeon
  • Hwang, Inchul
RU2792288C1
METHODS FOR UNDERSTANDING INCOMPLETE NATURAL LANGUAGE QUERY 2016
  • Sarikaya Ruhi
  • Derek Liu Xiaohu
RU2710966C2
METHOD AND SYSTEM FOR CONTROLLING DIALOGUE AGENT IN USER INTERACTION CHANNEL 2019
  • Moroz Darya Nikolaevna
  • Gureenkova Olga Aleksandrovna
  • Litinskij Aleksej Aleksandrovich
  • Pugin Pavel Yurevich
  • Burtsev Mikhail Sergeevich
RU2818036C1
METHOD FOR CONTROLLING A DIALOGUE AND NATURAL LANGUAGE RECOGNITION SYSTEM IN A PLATFORM OF VIRTUAL ASSISTANTS 2020
  • Ashmanov Stanislav Igorevich
  • Sukhachev Pavel Sergeevich
  • Zorkij Fedor Kirillovich
RU2759090C1
DETERMINING USER INTENT BASED ON ONTOLOGIES OF DOMAINS 2011
  • Chejer Adam Dzhon
  • Gudzzoni Did'E Rene
  • Brigem Kristofer Din
RU2541221C2
MAINTENANCE OF CONTEXTUAL INFORMATION BETWEEN USER INTERACTIONS WITH VOICE ASSISTANT 2018
  • Gruber, Thomas Robert
  • Cheyer, Adam John
  • Kittlaus, Dag
  • Guzzoni, Didier Rene
  • Brigham, Christopher Dean
RU2785950C2
USER INTENTION OUTPUT BASED ON PREVIOUS INTERACTIONS WITH VOICE ASSISTANT 2011
  • Gruber Tomas Robert
  • Chejer Adam Dzhon
  • Gudzzoni Did'E Rene
  • Brigem Kristofer Din
RU2544787C2
SERVICE ORCHESTRATION FOR INTELLIGENT AUTOMATED ASSISTANT 2011
  • Gruber Tomas Robert
  • Chejer Adam Dzhon
  • Gudzzoni Did'E Rene
  • Brigem Kristofer Din
RU2556416C2
DISAMBIGUATION BASED ON ACTIVE INPUT ELICITATION BY INTELLIGENT AUTOMATED ASSISTANT 2011
  • Gruber Tomas Robert
  • Chejer Adam Dzhon
  • Gudzzoni Did'E Rene
  • Brigem Kristofer Din
  • Sehddler Garri Dzhozef
RU2546605C2
ACTIVE INPUT REQUEST BY AUTOMATED INTELLIGENT ASSISTANT 2011
  • Gruber Tomas Robert
  • Chejer Adam Dzhon
  • Gudzzoni Did'E Rene
  • Brigem Kristofer Din
  • Dzhuli Richard Donal'D
  • Bastea-Forte Marchello
  • Sehddler Garri Dzhozef
RU2541208C2

RU 2 699 587 C2

Authors

Kannan Vishvak Sena

Uzelak, Aleksandar

Khvang Deniel Dzh.

Dates

2019-09-06Published

2016-01-15Filed