Data scientists from RTI gather around a large screen to look at a project.

RTI Center for Data Science

Practicing data science for social good

Data science is a relatively new and vibrant field that integrates novel and traditional sources of data in creative ways to solve problems and inform decision-making. As governments and foundations have become more data-driven, we have harnessed the methods and mindset of data science to solve social problems and inform public policy in new ways, and even to solve problems once thought to be intractable. We seek out projects that tackle important national problems, improve our local communities, and transform research. We work with federal agencies, foundations, and state and local jurisdictions on projects that span public and behavioral health, criminal and social justice, public safety, survey research, education, and the environment.

We approach data science through a combination of problem-solving acumen, human-centered design, advanced analytics, and agile software development. We embrace difficult problems and deliver creative, functional, and beautiful solutions. Our data scientists, software developers, and visual designers, in collaboration with our colleagues with expertise in multiple research domains, practice “tip-to-tail” data science; that is, we iterate with our clients to first define the problem and design solutions that fit their needs, then continue iterating to provide actionable insight through a combination of advanced analytics, product development, and modern reporting.  

Our vibrant data science team is a hub of innovation and collaboration, where people can thrive and be at their natural best while doing meaningful work to improve the world. Projects, presentations, publications, open-source applications, blogs, and media appearances demonstrate our thought leadership in the computational social sciences, most notably in data-driven justice. Along with our staff, our work has evolved from dashboards to deep learning, and from prototypes to fully deployed products. Inspired by RTI’s mission of improving the human condition, our Center is committed to practicing data science for social good.

Our technical capabilities include:

Advanced Analytics

  • Predictive modeling
  • Machine learning, artificial intelligence, neural networks, and deep learning
  • Microsimulation and agent-based modeling
  • Text analytics and natural language processing
  • Computer vision with images and video
  • Network analysis

Modern Reporting

  • Interactive data visualization
  • Dashboards and web applications
  • Data storytelling

Software Engineering

  • Data Engineering
    • Data pipelines with traditional and innovative ETL processes
    • Efficient data storage
    • System performance tuning
  • Cloud-based architectures designed to auto-scale
  • Agile software development
    • Rapid prototyping
    • Test-driven development
    • Continuous integration and continuous deployment
  • Data products from concept to prototype to production at-scale, integrating non-traditional sources, types, and amounts of data

Video

Detecting Physiological Changes Using Wearable Sensors

Challenges in Detecting Physiological Changes Using Wearable Sensor Data | SciPy 2019 | E. Preble
RTI Data Scientist Ed Preble presents his work on "Challenges in Detecting Physiological Changes Using Wearable Sensor Data" at the 2019 SciPy conference in Austin, Texas.

Video

Applied Natural Language Processing

Peter Baumgartner: Applied NLP: Lessons from the Field (spaCy IRL 2019)
In the most tweeted talk of the 2019 spaCy conference, RTI Data Scientist Peter Baumgartner discusses NLP’s distinct project management problems, and how to give your applied NLP projects the best chance of success

Video

Counting Arrest-Related Deaths

RTI International Data Scientist Peter Baumgartner - 2018 Tom Tom Founders Festival
RTI data scientist Peter Baumgartner talks about counting arrest-related deaths at the 2018 Tom Tom Festival. This project was conducted with funds awarded by the Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice. The views, opinions, and content expressed in this exhibition do not necessarily reflect the views, opinions, or policies of the Bureau of Justice Statistics (BJS), the Office of Justice Programs (OJP), or the U.S. Department of Justice (DOJ).

Video

Connected: A Social Network Analysis Tutorial with NetworkX

Rob Chew, Peter Baumgartner | Connected: A Social Network Analysis Tutorial with NetworkX
RTI Data Scientists Rob Chew and Peter Baumgartner at PyData Carolinas, providing a tutorial for using NetworkX to conduct social network analysis.

Video

rollmatch: An R Package for Rolling Entry Matching

rollmatch: An R Package for Rolling Entry Matching
RTI data scientist Kasey Jones discusses his R package for rolling entry matching (ROLLMATCH) at the July 2018 useR! conference in Brisbane, Australia.

News, Events, and Insights from the RTI Center for Data Science