Online event On demand In english
Documents processing is an inherent part of everyday work in basically every industry, as digitalized text documents such as contracts, web pages, forms, or emails are still the most common information carriers. Hence, many companies’ resources often engage in repetitive and highly time-consuming tasks including:
Luckily, the situation has changed fundamentally with the recent progress in AI, especially in Natural Language Processing. Sii is a part of this change. In Sii, we have built efficient and accurate AI pipelines that help our clients to automate many document processing challenges, leading to significant time savings and cost reduction.
During the webinar, we’d like to share our experience in this field demonstrating actual examples. Join us to learn how to replace labor-intensive, manual processes with sophisticated technology that will save your resources, money, and time.
Challenges in large-scale documents processing
AI-powered document reading pipeline
Key Information Extraction
Complex Documents Reading
Effective Search Engine
Bob Crutchfield has over 20 years of experience across multiple domains of Data and Analytics. In 10 years of software development with IBM, Bob led the development of restaurant back office systems - collecting and aggregating data from thousands of restaurants and millions of transactions on a nightly basis. Then over a 13-year career with Equifax, Bob led Data Strategy and Analytics for the Commercial Solutions business - including the development and deployment of the first small business credit scoring model leveraging machine learning. Bob later led a Product organization at Equifax in the development of a data and analytics platform for credit lenders to build, configure, and deploy their own credit scoring algorithms and decision strategies. Bob is now the SVP, Data Strategy and Operations for Dodge Construction Network - the leading provider of Commercial Construction data in the US. In this role, Bob is focused on solidifying DCN as a leader in Construction Data, powered by an intelligent Data Supply Chain that can automate tasks previously thought to be impossible to automate.
Marcin is an AI Architect with over ten years of experience collected in various machine learning projects, including natural language processing and computer vision. In his daily job, he converts the latest academic research into operating products and leads teams of AI experts. Marcin is currently working on an unstructured documents understanding and semantic search engine for the Polish language. Both projects touch on multiple areas of NLP, such as coreference resolution, unsupervised learning for sentence embeddings, document deduplication, etc. Previously, he built AI solutions for medical document verification and many computer vision projects, such as the perception module of an autonomous car, including object detection, visual object tracking, etc.
Over the past nine years, Jakub has worked on products based on Machine Learning, NLP, and Computer Vision as a team leader, researcher, and advisor. Recently, he has been developing solutions and Deep Learning models for automatic data extraction from documents and web pages, object detection and image classification, semantic search, and automatic question answering. He is co-author of an Open Source gesture recognition library for the Leap Motion device and author of the StratifiedGroupKFold method in a well-known Data Science tool called scikit-learn.
Czy chcesz opuścić tę stronę?