Dr. Ilkay Altintas, Chief Data Science Officer at the San Diego Supercomputer Center at the University of California, San Diego, was the invited speaker as part of the Distinguished Speakers Series, hosted by the Precision Institute at National University Nov. 30. The title of her talk was “Creating a Data Science Environment for Scientific, Societal, and Educational Impact.” This blog post is based on her talk.
The most important thing the fire fighters that have been battling the recent fast-moving wildfires destroying homes and blackening the landscape all the way from Temecula north to Ventura County want to know is where the blazes are headed next. If they know that, planes can drop water where it will do the most good. Firefighters can get out ahead and cut firebreaks. Authorities can advise residents in its path to evacuate. Lives and property can be saved.
California now has such a strategically powerful tool, thanks to the San Diego Supercomputer Center, at the University of California, San Diego, where Dr. Ilkay Altintas is the Chief Data Scientist. Dr. Altintas described the challenges of building what is called the WIFIRE system during the Distinguished Speakers Series lecture November 30 at the Precision Institute at National University. She also explained how similar uses of predictive analytics and data, which are being explored by the Institute, can be used to guide students to learning success.
Data to Improve Decision-making
“If you know where the fire is, where it’s headed, and at what rate of speed, you can take action and…manage the fire in a way that is less damaging,” she said.
The system gathers data on wind speed, humidity, the amount of brush available to fuel the fire, the steepness of mountain slopes, and other factors from many sources, including remote sensors, high-resolution images from strategically placed cameras, satellite information, and weather centers. All of that is fed into the supercomputer and it translates that information into maps that show the extent and spread of the fire. “The more information, the better the model, and the better the estimates of where it is going,” Dr. Altintas said. Deployment decisions can be made in a matter of minutes, instead of hours.
The Supercomputer Center is one of the partners helping the Precision Institute at National University research Precision Education, which is, in many ways, similar to the WIFIRE tool.
Like Fires, Learning is Complex
Like the movement of a wildfire, learning is a complex, dynamic, idiosyncratic process. Just as the WIFIRE tool gathers disparate data on fires, Precision Education will collect dozens of data points, such as previous performance, transcripts, past experiences, pre-test scores, time commitments, engagement, available learning supports, and much more about each student.
By analyzing the relationship between the data it collects and the movement of each fire, the WIFIRE tool can predict the movement of future fires. The Precision Education learning navigator, which is called NUNav, does something very similar with students and their learning. As students complete modules on the NUNav platform, it gathers data about their progress and combines that data with other student characteristics to create an algorithm determining the best future learning pathway. The more students who use the platform, the more data it collects, and the more precise its recommendations for learning pathways that are most likely to lead students to success.
The learning navigator can catch a student who is struggling before they fall too far behind and recommend evidence-based alternative ways to learn the material, ways that have worked for other students like them.
Data Science Has Many Different Applications
Dr. Altintas said data science is already being applied to better understand a variety of complex processes and also can be used successfully in education. But she also said “we need to be really cautious about how we do this.” The more disparate data points that are combined, the greater the risk of reaching unsupported conclusions, she said. In education, she said, data can be used to disadvantage particular groups of students or it can be used to help them.
But, she is optimistic about the potential of data science. “We are constantly improving what we know about the world, our environment, our systems, and our people and…by applying careful science and technology, we will get better as a society.”
To watch the full lecture: https://youtu.be/PIyvhyIWtIw
Precision Education at National University is a research-based initiative that is exploring new ways to leverage technology, open education resources, and predictive data analytics to adapt to student needs and guide them to successful completion of their academic and career goals. Learn more about the work being done through the Precision Institute at National University: https://www.nu.edu/precision