METHOD FOR PERFORMING TASKS Russian patent published in 2021 - IPC G06F17/00 

Abstract RU 2743898 C1

FIELD: computer engineering.
SUBSTANCE: invention relates to a method and a server for performing a specific type of task. The method consists in receiving, with the aid of a server, an indication of the first result of a task performed by an expert, wherein the quality assessment of the expert indicates the reliability of the first result, wherein the quality assessment of the expert is obtained from a database; the implementation of a machine learning algorithm (MLA) to accomplish the task of forming the second result of the task, wherein the quality assessment of the machine learning algorithm indicates the reliability of the second result, the quality assessment of the machine learning algorithm is obtained from the physical memory; determination of the confidence level parameter indicating the probability of the first result being correct, based on the first result, the second result, the quality assessment of the expert and the quality evaluation of the machine learning algorithm; determination of the probability exceedance, specified by a given confidence level parameter predefined by a threshold probability; and treating a task as completed with the first result if it is determined that the probability specified by a specified confidence level parameter, exceeds a predefined threshold probability.
EFFECT: increased efficiency with which the task is performed.

22 cl, 4 dwg

Similar patents RU2743898C1

Title Year Author Number
METHOD AND SYSTEM FOR DETERMINING RESULT OF TASK EXECUTION IN CROWDSOURCED ENVIRONMENT 2019
  • Fedorova Valentina Pavlovna
  • Gusev Gleb Gennadievich
  • Drutsa Alexey Valerievich
RU2744032C2
METHOD AND A SYSTEM FOR DETERMINING THE RESULT OF A TASK IN THE CROWDSOURCING ENVIRONMENT 2019
  • Drutsa Alexey Valerevich
RU2744038C2
METHOD AND SYSTEM FOR SELECTIVE PRESENTATION OF RECOMMENDATIONS IN A BROWSER APPLICATION 2018
  • Borisov Maksim Olegovich
  • Golikov Aleksey Vladimirovich
RU2739520C1
METHOD AND SYSTEM FOR DETERMINING THE NEXT STATE OF THE RECOMMENDATIONS BLOCK TO DISPLAY IN A WEB BROWSER 2018
  • Borisov Maksim Olegovich
  • Golikov Aleksey Vladimirovich
RU2746848C1
SYSTEM AND METHOD OF FORMING TRAINING SET FOR MACHINE LEARNING ALGORITHM 2017
  • Lakhman Konstantin Viktorovich
  • Chigorin Aleksandr Aleksandrovich
  • Yurchenko Viktor Sergeevich
RU2711125C2
METHOD AND SERVER FOR TRAINING MACHINE LEARNING ALGORITHM IN OBJECT RANKING 2020
  • Ustimenko Aleksej Ivanovich
RU2782502C1
TRAINING METHOD OF RATING MODULE USING THE TRAINING SELECTION WITH THE INTERFERENCE LABELS 2016
  • Gusev Gleb Gennadevich
  • Ustinovskij Yurij Mikhajlovich
  • Serdyukov Pavel Viktorovich
  • Fedorova Valentina Pavlovna
RU2632143C1
METHOD AND SERVER FOR REPEATED TRAINING OF MACHINE LEARNING ALGORITHM 2019
  • Pevtsov Sergey Evgenievich
  • Kostin Mikhail Yurievich
  • Chigin Anton Olegovich
  • Vasilyev Dmitry Sergeevich
RU2743932C2
METHOD AND SYSTEM OF RANKING OF A VARIETY OF DOCUMENTS ON THE PAGE OF SEARCH RESULTS 2017
  • Shishkin Alexandr Leonidovich
  • Vorobev Aleksandr Leonidovich
  • Gladkikh Ekaterina Anatolievna
RU2677380C2
METHOD OF ESTABLISHING TRAINING OBJECT FOR TRAINING MACHINE TRAINING ALGORITHM 2016
  • Gusev Gleb Gennadevich
  • Fedorova Valentina Pavlovna
  • Mishchenko Andrej Sergeevich
RU2637883C1

RU 2 743 898 C1

Authors

Fedorova Valentina Pavlovna

Gusev Gleb Gennadievich

Dates

2021-03-01Published

2018-11-16Filed