Executive Briefing: What you must know to build AI systems that understand natural language

New AI solutions in question answering, chatbots, structured data extraction, text generation, and inference all require deep understanding of the nuances of human language. David Talby outlines challenges, risks, and best practices for building natural language understanding (NLU)-based systems, drawing on examples and case studies from products and services built by Fortune 500 companies and startups over the past six years.

David also highlights some of the differences between language understanding and other machine learning and deep learning applications. You’ll learn about what gave NLU its reputation as an “AI complete” problem, the need to train domain-specific natural language processing (NLP) models for most systems because of the many languages we speak every day, the evolution from “traditional” machine learning and information retrieval techniques to current state-of-the-art systems, and guidelines to architecting a system that trains and serves large, current, and accurate domain-specific NLP models using open source software.