The National Science Foundation, an independent federal agency dedicated to promoting the progress of science, provided funding to three University of Arizona systems and industrial engineering researchers in September 2020. These prestigious grants, totaling more than $2.7 million, will allow the engineers to further their research into the fields of data science, STEM education and networked sensor systems.

"We're so proud of Ricardo, Qiang and Jianqiang for their hard work to earn these grants,” said Young-Jun Son, SIE department head. “Three new NSF awards in a department in one year would have been excellent news, but this is three in just one month.”

Machine Learning for Information Networks

Assistant professor Jianqiang Cheng is serving as co-principal investigator on a grant from NSF’s Information and Intelligent Systems division, which funds projects to study the interrelated roles of people, computers and information to increase the ability to understand data.

Yong Ge, an assistant professor of management information systems in the Eller College of Management, is the principal investigator on the project, “Learning to Hash Information Networks,” which will receive $330,000 in funding the first two years and up to $500,000 total. Many systems we use every day, such as social media networks, online shopping and internet search engines, involve information networks. One current focus in the field of data science is using machine learning to create effective representations of these networks. Doing so can lead to many benefits, like improving a network’s ability to offer accurate search results, rank the popularity of websites or predict the likelihood that two information nodes (such as two social media profiles) are related.

In this project, Cheng and Ge will be developing enhanced machine learning methods to improve these functions, while saving computational cost and storage space. The project aligns well with Cheng’s long-term research goal to develop more effective decision-making methods in industry.

“My role in the project is to develop technically sound and practically useful optimization methods for learning network data representations in a variety of information networks, including social networks and knowledge graphs,” Cheng said.

Increasing Kids’ STEM Engagement Through Sports

Professor Ricardo Valerdi is the founder of the Tucson-based nonprofit Science of Sport, which provides kids with hands-on sports activities to learn more about STEM topics. Kids learn about concepts like the launch angle of a baseball, the biomechanics of swimming or calculating the success rate of penalty kicks using fractions and decimals.

With a new $1.8 million grant from the NSF Advancing Informal STEM Learning program to increase youth engagement in STEM programs through sports, Valerdi will be building on the Science of Sport’s existing collaborations with Major League Baseball summer programs and Boys and Girls Clubs after school programs. In particular, his team will focus on providing greater exposure to STEM concepts for fourth through eighth grade Black, Latinx and female students – groups that are underrepresented in the field.

“This project will focus on expanding and refining Science of Baseball activities to incorporate growth mindset experiences that focus on the value of effort, determination and learning from mistakes in both athletics and STEM,” Valerdi said. “We will also study the enactment and outcomes of the program and develop a model for training informal STEM learning facilitators that includes both in-person and online components.”

Valerdi and his team, including Erin Turner, a professor in the College of Education, will use this funding to gather empirical evidence about what works and what doesn’t so similar sports-focused STEM programming can be developed elsewhere.

Managing Battery Life in Networked Sensor Systems

SIE assistant professor Qiang Zhou is the principal investigator of a $479,000 grant from the NSF Division of Civil, Mechanical and Manufacturing Innovation to better manage the battery life of wireless devices in networked sensor systems. In these systems, hundreds or thousands of battery-powered nodes collect data and share it with the other sensors

These networks have been used for applications including efficient water quality monitoring, forest fire detection and machine health monitoring for many years. With the rise of the Internet of Things, these networks have become critical components of “smart” devices, ranging from cellphones to refrigerators to entire cities.

The wireless nature of these systems means they are flexible, but the limited nature of battery life poses challenges. This project will investigate ways to improve the performance of entire networked sensor systems by more effectively monitoring the life of individual batteries and developing cost-effective group replacement policies. It will involve creating an algorithm to accurately predict remaining battery life and developing a policy and schedule for maintaining and replacing groups of batteries.

“For IoT-enabled smart homes and cities, smart manufacturing, smart energy, and smart agriculture, an unprecedented number of battery-powered devices such as wireless sensors will be deployed in the field,” Zhou said. “Reducing the battery maintenance cost will be the key to their success.”