Knowledge Graph & NLP

Schedule:

6:00 - Doors & Food
6:30 - Talk 1
7:15 - Talk 2
8:00 - Wrap & Chat

Talk 1: Building a knowledge graph at Thomson Reuters
Speakers:
Dan Bennett, VP, Enterprise Data Services @ Thomson Reuters
Yoni Mataraso, Software Development Manager and Product Owner @ Thomson Reuters

Abstract:
Just a few years ago a knowledge graph was the domain of academic papers, today they underpin the natural language capabilities of Alexa, Siri, Cortana and Google Now. Graphs are a natural fit for this use case: treating every data item as equivalent and embracing rapid schema mutation. For the past few years, Thomson Reuters has been building a professional information knowledge graph to power our next generation of products. Our graph is RDF based, fast growing and supports a number of different products and user experiences. In this session, Dan & Yoni will cover our experiences, architecture, tools and lessons learned from building, integrating and maintaining a 100bn triple graph.

Talk 2: Natural language processing in healthcare
Speaker:
Benjamin Bernstein, Machine Learning Engineer @ Roam Analytics

Abstract:
Roam Analytics is a startup focused on healthcare and NLP that extracts information locked away in raw text like outcomes, complex provider decision making, and patient perceptions. In this talk, I'll introduce the challenges of working with text data and describe our approach to ingesting and transforming for analysis. Also, we'll explore creating efficient analysis pipelines and discuss how that enables us to do highly effective concept extraction with machine learning. At the end, we'll open it up for questions and discussion about healthcare and NLP.