Processing

Please wait...

PATENTSCOPE will be unavailable a few hours for maintenance reason on Tuesday 25.01.2022 at 12:00 PM CET
Settings

Settings

Goto Application

1. WO2022010731 - COMPACT REPRESENTATION AND TIME SERIES SEGMENT RETRIEVAL THROUGH DEEP LEARNING

Publication Number WO/2022/010731
Publication Date 13.01.2022
International Application No. PCT/US2021/040081
International Filing Date 01.07.2021
IPC
G06F 16/28 2019.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
20of structured data, e.g. relational data
28Databases characterised by their database models, e.g. relational or object models
G06F 16/25 2019.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
20of structured data, e.g. relational data
25Integrating or interfacing systems involving database management systems
G06F 16/22 2019.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
20of structured data, e.g. relational data
22Indexing; Data structures therefor; Storage structures
G06F 16/2458 2019.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
20of structured data, e.g. relational data
24Querying
245Query processing
2458Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
G06N 3/08 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
G06N 3/04 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
Applicants
  • NEC LABORATORIES AMERICA, INC. [US]/[US]
Inventors
  • MIZOGUCHI, Takehiko
  • SONG, Dongjin
  • CHEN, Yuncong
  • LUMEZANU, Cristian
  • CHEN, Haifeng
Agents
  • BITETTO, James J.
Priority Data
17/364,12530.06.2021US
63/048,68007.07.2020US
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) COMPACT REPRESENTATION AND TIME SERIES SEGMENT RETRIEVAL THROUGH DEEP LEARNING
(FR) REPRÉSENTATION COMPACTE ET RÉCUPÉRATION DE SEGMENTS DE SÉRIE CHRONOLOGIQUE PAR APPRENTISSAGE PROFOND
Abstract
(EN) Systems and methods for retrieving similar multivariate time series segments are provided. The systems and methods include extracting (920) a long feature vector and a short feature vector from a time series segment, converting (930) the long feature vector into a long binary code, and converting (930) the short feature vector into a short binary code. The systems and methods further include obtaining (940) a subset of long binary codes from a binary dictionary storing dictionary long codes based on the short binary codes, and calculating (950) similarity measure for each pair of the long feature vector with each dictionary long code. The systems and methods further include identifying (960) a predetermined number of dictionary long codes having the similarity measures indicting a closest relationship between the long binary codes and dictionary long codes, and retrieving (970) a predetermined number of time series segments associated with the predetermined number of dictionary long codes.
(FR) L'invention concerne des systèmes et des procédés de récupération de segments de série chronologique à plusieurs variables similaires. Les systèmes et les procédés comprennent l'extraction (920) d'un vecteur de caractéristique long et d'un vecteur de caractéristique court à partir d'un segment de série chronologique, la conversion (930) du vecteur de caractéristique long en un code binaire long, et la conversion (930) du vecteur de caractéristique court en un code binaire court. Les systèmes et les procédés comprennent en outre l'obtention (940) d'un sous-ensemble de codes binaires longs à partir d'un dictionnaire binaire stockant des codes longs de dictionnaire sur la base des codes binaires courts, et le calcul (950) d'une mesure de similarité pour chaque paire du vecteur de caractéristique long avec chaque code long de dictionnaire. Les systèmes et les procédés comprennent en outre l'identification (960) d'un nombre prédéterminé de codes longs de dictionnaire ayant les mesures de similarité indiquant la relation la plus proche entre les codes binaires longs et les codes longs de dictionnaire, et la récupération (970) d'un nombre prédéterminé de segments de série chronologique associés au nombre prédéterminé de codes longs de dictionnaire.
Related patent documents
Latest bibliographic data on file with the International Bureau