Several RTI researchers will be presenting during the annual SciPy2019 Conference, held July 8-14 in Austin, Texas.
Wearable sensors offer the potential to provide real-time health status to individuals, but there are many challenges, such as variation in use during a normal day. Ed Preble, Kristin Gilchrist, and Meghan Hegarty-Carver’s talk “Challenges in Detecting Physiological Changes Using Wearable Sensor Data,” discusses how to process massive variations in wearable data to create visualizations and identify useful patterns.
Large labeled data sets are a prime requirement for machine learning adoption. RTI’s Rob Chew and Michael Wenger will present a solution to address that. “SMART: An Open Source Data Labeling Platform for Supervised Learning” will discuss an open source application to help teams efficiently build labeled training data sets for supervised machine learning tasks. SMART offers an interface for creating labeled data sets, supports active learning to help reduce the required amount of labeled data, and incorporates inter-rater reliability statistics to provide insight into label quality.
SciPy brings together more than 800 people from industry, academia, and government to learn, discuss projects, and collaborate on code development.