Processing

Please wait...

Settings

Settings

Goto Application

1. WO2019224629 - TRAINING DATA EXPANSION FOR NATURAL LANGUAGE CLASSIFICATION

Publication Number WO/2019/224629
Publication Date 28.11.2019
International Application No. PCT/IB2019/053517
International Filing Date 30.04.2019
IPC
G06F 17/27 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
20Handling natural language data
27Automatic analysis, e.g. parsing, orthograph correction
CPC
G06F 40/211
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
20Natural language analysis
205Parsing
211Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
G06F 40/226
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
20Natural language analysis
205Parsing
226Validation
G06F 40/284
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
20Natural language analysis
279Recognition of textual entities
284Lexical analysis, e.g. tokenisation or collocates
G06F 40/30
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
40Handling natural language data
30Semantic analysis
G06K 9/6256
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting
H04L 51/02
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
51Arrangements for user-to-user messaging in packet-switching networks, e.g. e-mail or instant messages
02with automatic reactions or user delegation, e.g. automatic replies or chatbot
Applicants
  • INTERNATIONAL BUSINESS MACHINES CORPORATION [US]/[US]
  • IBM (CHINA) INVESTMENT COMPANY LTD. [CN]/[CN] (MG)
  • IBM DEUTSCHLAND GMBH [DE]/[DE] (MG)
Inventors
  • CRUDELE, Michele
  • PERRONE, Antonio
Agents
  • KUISMA, Sirpa
Priority Data
15/988,29524.05.2018US
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) TRAINING DATA EXPANSION FOR NATURAL LANGUAGE CLASSIFICATION
(FR) EXPANSION DE DONNÉES D'APPRENTISSAGE POUR UNE CLASSIFICATION DE LANGAGE NATUREL
Abstract
(EN) A computer-implemented method for training a natural language classifier associated with a chat interface of a computer system is provided. The method may include receiving a training dataset comprising an initial set of expressions corresponding to an intent. Additional expressions corresponding to the intent may be generated, wherein the additional expressions are generated based on the initial set of expressions corresponding to the intent. The natural language classifier may be trained based on the initial set of expressions and the additional expressions corresponding to the intent. The trained natural language classifier may be implemented to determine an intent expressed by a detected query based on the initial set of expressions and the additional expressions by which the natural language classifier was trained.
(FR) L'invention concerne un procédé mis en œuvre par ordinateur pour entraîner un classificateur de langage naturel associé à une interface de discussion d'un système informatique. Le procédé peut comprendre la réception d'un ensemble de données d'apprentissage comprenant un ensemble initial d'expressions correspondant à une intention. Des expressions supplémentaires correspondant à l'intention peuvent être générées, les expressions supplémentaires étant générées sur la base de l'ensemble initial d'expressions correspondant à l'intention. Le classificateur de langage naturel peut être entraîné sur la base de l'ensemble initial d'expressions et des expressions supplémentaires correspondant à l'intention. Le classificateur de langage naturel entraîné peut être implémenté pour déterminer une intention exprimée par une interrogation détectée sur la base de l'ensemble initial d'expressions et des expressions supplémentaires ayant entraîné le classificateur de langage naturel.
Latest bibliographic data on file with the International Bureau