Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020139635 - AUTOMATED TRAINING AND SELECTION OF MODELS FOR DOCUMENT ANALYSIS

Publication Number WO/2020/139635
Publication Date 02.07.2020
International Application No. PCT/US2019/066986
International Filing Date 17.12.2019
IPC
G06F 16/383 2019.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
16Information retrieval; Database structures therefor; File system structures therefor
30of unstructured textual data
38Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
383using metadata automatically derived from the content
CPC
G06K 9/00469
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
00442Document analysis and understanding; Document recognition
00469Document understanding by extracting the logical structure, e.g. chapters, sections, columns, titles, paragraphs, captions, page number, and identifying its elements, e.g. author, keywords, ZIP code, money amount
G06K 9/6218
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
6218Clustering techniques
G06N 20/20
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
20Ensemble learning
G06N 5/003
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computer systems using knowledge-based models
003Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
G06N 5/041
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computer systems using knowledge-based models
04Inference methods or devices
041Abduction
Applicants
  • ICERTIS, INC. [US]/[US]
Inventors
  • CHAUDHARI, Dhruv
  • SHAH, Harshil
  • JAIN, Amitabh
  • DARDA, Monish, Mangalkumar
Agents
  • BRANCH, John W.
Priority Data
16/231,84224.12.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) AUTOMATED TRAINING AND SELECTION OF MODELS FOR DOCUMENT ANALYSIS
(FR) ENTRAÎNEMENT ET SÉLECTION AUTOMATISÉS DE MODÈLES POUR L'ANALYSE DE DOCUMENTS
Abstract
(EN)
Embodiments are directed to a machine learning engine that determines training documents and validation documents from a plurality of documents. The machine learning engine may determine attributes associated with the documents. In response to receiving a request to predict attribute values of a selected document the machine learning engine may train a plurality of ML models to predict the attribute values based on the training documents and the attributes and associate the trained ML models with an accuracy score. The machine learning engine may determine candidate ML models from the trained ML models based on the training accuracy scores. The machine learning engine may evaluate and rank the candidate ML models based on the request and the validation documents. The machine learning engine may generate confirmed ML models based on the ranked candidate ML models such that the confirmed ML models may answer the request.
(FR)
Selon des modes de réalisation, l'invention concerne un moteur d'apprentissage automatique qui détermine des documents d'entraînement et des documents de validation à partir d'une pluralité de documents. Le moteur d'apprentissage automatique peut déterminer des attributs associés aux documents. En réponse à la réception d'une demande de prédiction de valeurs d'attribut d'un document sélectionné, le moteur d'apprentissage automatique peut entraîner une pluralité de modèles de ML pour prédire les valeurs d'attribut en fonction des documents d'entraînement et des attributs et associer les modèles de ML entraînés à un score de précision. Le moteur d'apprentissage automatique peut déterminer des modèles de ML candidats à partir des modèles de ML entraînés en fonction des scores de précision d'entraînement. Le moteur d'apprentissage automatique peut évaluer et classer les modèles de ML candidats en fonction de la demande et des documents de validation. Le moteur d'apprentissage automatique peut produire des modèles de ML confirmés en fonction des modèles de ML candidats classés de sorte que les modèles de ML confirmés puissent répondre à la demande.
Also published as
Latest bibliographic data on file with the International Bureau