METHOD FOR OBTAINING LOW-DIMENSIONAL NUMERIC REPRESENTATIONS OF SEQUENCES OF EVENTS Russian patent published in 2021 - IPC G06F17/00 G06N3/00 G06N3/06 G06Q20/04 

Abstract RU 2741742 C1

FIELD: physics.

SUBSTANCE: invention relates to the field of information technologies, in particular to a method of obtaining low-dimensional numerical representation of sequences of events. Disclosed is a computer-implemented method of obtaining low-dimensional numerical representation of sequences of events, comprising steps of: obtaining a set of input data, characterizing events, aggregated in a sequence and associated with at least one information entity, wherein pre-processing of said set of input data, in which: generating positive and negative pairs of sequences of transaction events, using a transactional event encoder generating a vector representation of each transaction event from said set of attributes, normalizing numerical variables; processing time marks, concatenating the obtained vector representations of categorical variables and normalized numerical variables; single event vector is generated based on the resultant concatenation.

EFFECT: higher efficiency of generating features for machine learning models by generating low-dimensional numerical representations of sequences of events.

7 cl, 8 dwg, 8 tbl

Similar patents RU2741742C1

Title Year Author Number
METHOD OF CALCULATING CLIENT CREDIT RATING 2019
  • Babaev Dmitrij Leonidovich
  • Umerenkov Dmitrij Evgenevich
  • Savchenko Maksim Sergeevich
RU2723448C1
NEURAL NETWORK TRANSFER OF THE FACIAL EXPRESSION AND POSITION OF THE HEAD USING HIDDEN POSITION DESCRIPTORS 2020
  • Burkov Egor Andreevich
  • Pasechnik Igor Igorevich
  • Lempitsky Victor Sergeevich
RU2755396C1
TRAINING GAN (GENERATIVE ADVERSARIAL NETWORKS) TO CREATE PIXEL-BY-PIXEL ANNOTATION 2019
  • Galeev Danil Fanilievich
  • Sofiyuk Konstantin Sergeevich
  • Rukhovich Danila Dmitrievich
  • Konushin Anton Sergeevich
  • Romanov Mikhail Viktorovich
RU2735148C1
MODELLING PEOPLE'S CLOTHING BASED ON MULTIPLE POINTS 2021
  • Grigorev Artur Andreevich
  • Lempitsky Victor Sergeevich
  • Zakharkin Ilya Dmitrievich
  • Mazur Kirill Yevgenevich
RU2776825C1
METHOD OF TRAINED RECURRENT NEURAL NETWORK DEBUGGING 2019
  • Zharov Yaroslav Maksimovich
  • Korzhenkov Denis Mikhajlovich
RU2715024C1
METHOD FOR CONSTRUCTING A DEPTH MAP FROM A PAIR OF IMAGES 2022
  • Maslovich Nikolaj Romanovich
  • Yashunin Dmitrij Aleksandrovich
  • Derendyaev Ilya Vasilevich
RU2806009C2
METHOD FOR CREATING COMBINED NEURAL NETWORK CASCADES WITH COMMON FEATURE EXTRACTION LAYERS AND WITH MULTIPLE OUTPUTS, TRAINED ON DIFFERENT DATASETS SIMULTANEOUSLY 2021
  • Levashov Aleksej Evgenevich
  • Dolgov Vasilij Sergeevich
  • Erpylov Aleksej Anatolevich
RU2779408C1
METHOD AND SYSTEM FOR VIDEO SCENES SEGMENTATION 2022
  • Leksutin Roman Valerevich
  • Zhilin Evgenij Yurevich
RU2783632C1
METHOD OF TRAINING DEEP NEURAL NETWORKS BASED ON DISTRIBUTIONS OF PAIRWISE SIMILARITY MEASURES 2016
  • Ustinova Evgeniya Sergeevna
  • Lempitskij Viktor Sergeevich
RU2641447C1
METHOD FOR PROVIDING COMPUTER VISION 2022
  • Rukhovich Danila Dmitrievich
  • Konushin Anton Sergeevich
  • Konushin Anton Sergeevich
RU2791587C1

RU 2 741 742 C1

Authors

Babaev Dmitrij Leonidovich

Ovsov Nikita Pavlovich

Kireev Ivan Aleksandrovich

Dates

2021-01-28Published

2020-02-14Filed