Presented By
O’Reilly + Cloudera
Make Data Work
March 25-28, 2019
San Francisco, CA

Applied AI and NLP for enterprise contract intelligence (sponsored by ThoughtTrace)

Joel Hron (ThoughtTrace), Nick Vandivere (ThoughtTrace)
4:20pm5:00pm Wednesday, March 27, 2019
Sponsored
Location: 2022
Average rating: ****.
(4.00, 1 rating)

What you'll learn

  • Learn ThoughtTrace's approach to building and deploying enterprise-ready AI SaaS software

Description

Joel Hron and Nick Vandivere walk you through ThoughtTrace’s journey, highlighting its beginnings as a company and sharing the challenging use cases the company tackled first.

In the United States, companies obtain the right to explore for and extract minerals through direct negotiation with individual mineral owners. As you might imagine, the contracts executed to produce these minerals contain a variety of provisions that determine a company’s rights and obligations. In the US, this problem is exceptionally challenging given that a company must come to terms on a unique agreement with each individual mineral owner, which results in a single company managing tens of thousands or in some cases hundreds of thousands of leases. These leases may span up to 100 years in time with varying degrees of complexity, ranging from two-page form documents to 50 or more pages of legal obligation. The image qualities and varying representations of the language present unique challenges to building high-functioning AI solutions.

The domain specificity of this use case prohibits ThoughtTrace from defaulting to many of the state-of-the-art deep learning techniques. The company faces the cold start problem—building models with little to no prelabeled public data—and is often tasked with inferring over 250 unique classes in a multilabel classification problem. As a result, it uses a variety of techniques and technologies, including embeddings representations, bag of words representations, active learning, machine learning, and more traditional NLP approaches to address the various elements of this problem.

The human factor challenges can often be more challenging than even the technical. Given an AI model of the contract, presenting the results to a nontechnical individual in an intuitive way is exceptionally challenging. This requires a tremendous amount of effort in all facets of the business, from building models with high accuracy to designing a UX that facilitates simple action and feedback to exceptional customer engagement and education.

This session is sponsored by ThoughtTrace.

Photo of Joel Hron

Joel Hron

ThoughtTrace

Joel Hron is chief technology officer at ThoughtTrace, where he’s responsible for product innovation and strategy, the development and application of AI and machine learning, and full stack software development. His team is comprised of data scientists, full stack developers, DevOps engineers, quality assurance engineers, and domain experts who all work together to deliver innovative applied AI solutions that function in a highly scalable and reliable way. Previously, he held various engineering and management roles at Anadarko Petroleum Corporation, where he helped develop and shepherd new technologies enabling digital operations, better reservoir characterization, and field development optimization, with a focus on applied data science, artificial intelligence, machine learning, and other big data technologies. He holds a BS in mechanical engineering from TCU and an MS in mechanical engineering from the University of Texas.

Photo of Nick Vandivere

Nick Vandivere

ThoughtTrace

Nick Vandivere is CEO at ThoughtTrace, where he has led the company’s transformation into a technology and thought leader for the application of applied artificial intelligence and machine learning. He believes that product innovation, reliability, and an outstanding user experience are the required ingredients for exceptional performance and long-term value. Previously, he was an officer in the US Army and served in an advisory role with the US Department of State. Nick holds a BS in economics from Texas A&M University.