Leveraging deep learning techniques to unlock key information from complex documents
AI/Machine Learning
• November 2021
Leveraging deep learning techniques to unlock key information from complex documents
About
Leveraging deep learning techniques to unlock key information from complex documents
About
Spesso le informazioni rilevanti si presentano in forma non strutturata all'interno di documenti e testi. I progressi nel campo del deep learning hanno portato allo sviluppo di modelli sempre più avanzati che costituiscono la chiave di accesso a queste informazioni. Dotarsi di tecnologie per computer vision e NLU rappresenta oggi una necessità per le aziende che vogliono ottimizzare e automatizzare i loro processi di business. In questa sessione vedremo insieme alcune tecniche utilizzate nel contesto della intelligent document processing per l'estrazione di informazioni da documenti complessi.
Consulta il materiale degli speakers:
https://drive.google.com/file/d/1zBF9sL4xs2Z4fsFyA9Qz5TUI9XMUze43/view?usp=sharing
Language
Italian
Level
Beginner
Length
32 minutes
Type
online conference
About the speakers
About the speakers
Francesco Visalli
Principal Research Scientist, R&D Team Leader •
Altilia
Research scientist with 5+ years of experience in the fields of Natural Language Processing and Understanding, Machine Reading Comprehension, Question Answering, Document Layout Analysis and Understanding, IR, IE, Computer Vision, OCR.
Since April 2021 he leads a team of 10 researchers and innovators, dealing with the SoTA of Deep Learning algorithms to industrialize innovative solutions of intelligent document processing and decision intelligence. Francesco takes part to the scientific community, contributing with several papers.
Antonio Patrizio
Research Scientist, R&D Tech Lead •
Altilia
Antonio is a member of Altilia since 2019 and tech lead of the R&D team since July 2021. His main focus areas are Natural Language Processing and Understanding, Computer Vision, Document Layout Analysis and Understanding, Information Retrieval, Text Detection and Recognition. Antonio published a paper called "A Two Step Fine-tuning Approach for Text Recognition on Identity Documents" that talks about a method for extracting key information from dirty scanned documents.
Details
Language
Italian
Level
Beginner
Length
32 minutes
Type
online conference