EJB Talks Podcast

Gavin Rozzi

Using Data and Emerging Technologies to Improve Urban Life

May 11, 2020

The Rutgers Urban and Civic Informatics Lab (RUCI Lab) was recently founded at the Bloustein School, with the mission of using new sources of data and emerging technologies to study the health of cities, and by doing so, improve the quality of urban life. This week, Stuart Shapiro talks with the Bloustein School’s newest employee, Gavin Rozzi, who joined the school as a Research Computing Specialist for the RUCI Lab just days before the state’s stay-at-home order took effect. They discuss some of the projects the RUCI Lab will be working on, including using geographic and social media data to predict the long-term social and economic impacts of COVID-19 on society, as well as Gavin’s personal interest utilizing emerging technology — 3D printing PPEs for frontline healthcare workers.

Stuart Shapiro
Welcome to another episode of EJB talks. I’m Stuart Shapiro, the Associate Dean of Faculty at the Bloustein School. And the purpose of this podcast is to talk with my colleagues about issues affecting people in New Jersey, the United States and the world. Today we’re going to talk with Bloustein’s newest employee, Gavin Rozzi, and discuss technology and its role in combating and understanding COVID-19. Welcome, Gavin.

Gavin Rozzi
Thanks for having me today. Stuart. It’s a pleasure to be here.

Stuart Shapiro
Gavin, you joined us literally a day or two before we went remote. You managed to be in the building I think for a day? Can you tell us a little bit about your background and what brought you to Bloustein?

Gavin Rozzi
Yes, certainly. And that was definitely an interesting way to start. I was in my office for about a day and a half before we went remote, and now here we are. So I came to Bloustein from southern New Jersey, doing both my undergraduate degree and my master’s at Stockton University in Galloway. I’m currently studying data science and strategic analytics. My background has been working for our Public Policy Research Center at Stockton, where I was one of the co-authors on our report on the underground construction economy in New Jersey. We looked at employer misclassification and other shady dealings within the construction industry as a whole, and how our policymakers can come up with a response to that.

In addition to that, I’ve been doing a good deal of work with technology on my own. In my past work, I’ve done consulting for private industry and New Jersey municipal governments. I’ve done a lot of work with web development and emerging technologies; one of my biggest projects has been developing a statewide system to make the process of public records requests paperless, and also warehouse and archive that data for research so we can understand and inform this ongoing policy debate about the public records laws of New Jersey.

Stuart Shapiro
That’s great. OPRA laws are incredibly valuable. I know. I’ve made some requests and my wife, who’s a school board member, has been on the receiving end of them. At the federal level the FOIA laws — Is there anything like that for FOIA, what you’re doing?

Gavin Rozzi
Yes, there is. Although the landscape with FOIA is a little bit more complex. So with OPRA under current law in New Jersey, it’s typically within seven business days that you’ll get a response. FOIA kicks that up to 20 days. I actually explored the possibility of maybe doing something like this at the federal level. But the issue is, a lot of federal agencies, they’re adopting their own proprietary portals. Like for example, the FBI, you used to be able to send them an email or fax in a request. They now require you to use their specific websites and portals. And that’s not really conducive for gathering metadata about the requests or even archiving the records. So it’s tough to do in an automated fashion. But there are some platforms out there. Like, for example, MuckRock, you know. They’ve built custom applications for handling a lot of this stuff. But the challenge is, especially as you see some trends at the federal level, as you see some of this administrative rulemaking taking place by some of these agencies, they’re starting to be more restrictive, in terms of the manner that they’re accepting FOIA requests. So it’s kind of this ebb and flow that we’re seeing with policy at the federal level with respect to public records compliance.

Stuart Shapiro
Yes, it’s interesting stuff. We could probably do a whole episode on FOIA. I used to work for the federal government, so I have been on the receiving end of those requests, and I certainly learned how to make as little of them as possible. But we can chat about that some other time. Before we get into your work at Bloustein, I have to ask about your recent endeavor, printing 3D masks. This got some news coverage. Where did you get the idea? And can you describe the work you put into it?

Gavin Rozzi
Yes, certainly. So I recently got into 3D printing myself. I purchased my first 3D printer over the holiday season. It was just something that I’ve always been interested in learning about, you know, emerging technologies. As it begins to be adopted more, you start to see some pretty rapid progress in terms of the hardware and software. So, you know, as we all started working remotely and being at home, I looked at my 3D printer, I saw it was just sitting there idle. And, I was thinking about some of these really tough situations that we have to confront, both as a state and as a nation, where you have this shortage of personal protective equipment for a lot of these frontline health care workers.

Gavin Rozzi
I certainly wasn’t the first person to start doing this, but I read about this being done in other states and I saw a shortage. I saw a real opportunity for this technology to make a difference. I started looking at some of these different designs online, seeing what can be used to fill this need. And I’ve been printing things around the clock. I’ve been having healthcare workers pick the materials up, and I’m making I’ve been mailing some stuff out. The work that goes into this, the process of 3D printing objects typically begins with either designing or obtaining the design files for a 3D object. And it’s really amazing when you think about it, we can now take physical objects and represent them as digital data and send this data around the world. Anybody could improve upon these designs and modify them, and using that, people can print out copies of masks, straps for the masks for comfort. And there’s a lot of variables at play. There are different types of materials you can work with. There are a lot of different settings that need to be adjusted in terms of the parameters of the printer, how fast it’s printing, temperatures of the bed, and the like. So it really was a lot of trial and error.

What people don’t see is the behind the scenes work of troubleshooting a misaligned print bed, dealing with a lot of these roadblocks that come up to get the printers working. But it’s really been a valuable experience because I’ve heard from people that are working at the COVID testing sites that are using my stuff. I’ve heard from people that are in these units at hospitals where they really can’t get anything. And being able to make this difference, it’s certainly the least I can do. I’m happy to be able to give back professionally. It’s very important to me to be able to take an emerging technology like this and use it to make a positive impact on people’s lives.

Stuart Shapiro
That’s great! About how many do you think you’ve printed so far?

Gavin Rozzi
So far, I would say I’ve made probably at least 200 masks, face shields, and straps combined. Initially, it was taking me a little bit longer because I only had one printer, but I acquired a second printer. And I switched to more efficient designs that were able to cut the time down from about four or five hours per mask to only about an hour and a half to two, just by virtue of having a more efficient design that allows the printer to construct the mask in a shorter timeframe.

Stuart Shapiro
Well, we should get to your work here at Bloustein; that’s another subject I’d love to learn more about. You’re working with Dean Thakuriah on our new Rutgers Urban and Civic Informatics Lab or as we’re calling it, the RUCI Lab. Can you explain to our listeners what the lab is and what it hopes to accomplish?

Gavin Rozzi
Yes, certainly. So, Dean Thakuriah brings a lot of interesting things to the table with her past work with the Urban Big Data Center and her very strong network of international collaborators and colleagues. We’re going to be drawing upon that to execute the vision of the RUCI Lab. We’re going to be supporting socially just, environmentally resilient, and healthy cities through big data analytics and emerging technology. So really, what the big picture of that is, is using data and doing data science to support data-driven decision making, to really work with a lot of the resources that we have within Bloustein and within Rutgers. You know, we have a lot of very strong graduate programs at Bloustein be it the MPI public informatics, urban planning, etc. And our vision is for this lab to serve as a resource for those students in those programs.

And to that end, we’re going to be building out a big data infrastructure. We’re going to be compiling a lot of very important data sets that when combined together, can be used to study the health of cities, study how new sources of data and a lot of these critical issues can be used to ultimately improve the quality of life within cities, and give our decision-makers and partners within government and within private industry, new insights based upon data.

Stuart Shapiro
That’s great. I was going to ask you to pick out an example but let’s focus on one in particular here that I know we’re talking about on our website, the CES-COVID-19 project, which specifically deals with the crisis we’re in right now. What’s the idea behind this project?

Gavin Rozzi
So with CES-COVID-19, the big picture goal with this initiative is a lot of the data that we are looking at within the media and academically, in research, deals with the public health and containment aspect of COVID-19. And I don’t mean to discount that, that’s certainly very important. And those efforts need to continue. But what the conversation has not centered on as much is, what are the long term social and economic impacts of COVID-19 going to be to society? We need to look at what these big picture issues are going to be because obviously, the socioeconomic damage as a result of the pandemic has been a really significant cause for concern. So the RUCI Lab’s response to that is to develop a cyberinfrastructure that’s going to be capturing real-time or close to real-time data on these effects. And we’re going to be doing this by combining different sources of data. So we’re looking at geographic data, census data. I think a really interesting source of insight is going to be the social media data that we’re collecting.

You know, we’ve all seen Twitter, we’ve all seen all this stuff that’s being put out on social media. But looking at it from the big picture of collecting tweets to analyze trends within cities, you know–how people are doing, how policymakers and decision-makers are responding to these issues. So ultimately, I believe that by bringing together these sources of data, it’s going to be a resource for researchers that are going to be answering some of these big questions that we’re going to need to address as a society. But also, I think it’s going to be a really useful tool for some of our graduate students, giving students the ability to get into the mindset of data-driven decision making, and using real-world data to study these very complex phenomena that we’re looking at. So ultimately, we’re going to be using data to answer some of these big questions that COVID-19 is causing us to have to address as a society.

Stuart Shapiro
Let me ask you this perhaps the biggest question we’re facing right now. Certainly if one reads the media, it’s the question getting the most attention is when and how should we reopen up the economy, given the public health risks and the risks of economic dislocation? Will your data be able to speak to that question at all? Do you think?

Gavin Rozzi
I think it’s definitely going to be able to play a role in that. And I think, you know, with the conversation about opening up again, I was very pleased to see that Bloustein and Rutgers are being represented on the Governor’s Task Force for that. So that’s certainly a good thing for moving that dialogue forward. In terms of when and how to open up the economy, we’re looking at a few of these different data sources. This isn’t going to be on the level of something like a domain-specific public health analysis. But I think in terms of addressing issues like the damage to housing markets, issues associated with transportation, and going through cities, I think what our data set is really going to be able to play a significant role in is addressing some of the systemic inequities that have been aggravated by COVID-19. So I think a big part of that conversation is going to need to be, are there existing systems in place that have amplified inequities, you know, in poverty–along the lines of gender along the lines of race and other disparities? And I think one key insight that we’ll be able to deliver is, you know, which areas are really suffering even further as a result. And if we can get that information to our policymakers and decision-makers, hopefully it can allow them to craft tailored policy solutions that can kind of address these inequities and put a bandage on it, rather than aggravating it further. Because, you know, it’s been said before, we don’t want the cure to be worse than the disease itself.

Stuart Shapiro
Right. Do you have data on people’s movements? I know there’s been a lot of talk that what we need to do is test and then trace movements and such. Is that part of your data gathering or is that something that we don’t have access to?

Gavin Rozzi
I would say yes and no. Yes, to the extent that with the social media data that we’ve been collecting, we’re particularly interested in geotagged data. You know, you can get tweets like for example, we can request Twitter data from within certain cities, within certain geographic locations, within the tri-state area. So using those coordinates and other GIS data, we can certainly track movements. Another really important source of data that we’re looking at is transit data. For example, New Jersey Transit, other rail systems, bus systems, we can see when those trains are arriving, whether they’re running late, and pulling in all those different feeds over time. And combined with other data that can kind of give us some insight into how transit is playing a role in both the spread of the disease and the pandemic. But also, you know, for urban living as a whole in those larger conversations.

We’re certainly interested as we develop the cyberinfrastructure capabilities of the lab further, we’re very interested in partnering with other outside agencies and other stakeholders to maybe take a look at getting some restricted data sets; you know, some more sensitive information that could be used in research, you know, and obviously, that comes along with the need to address the cybersecurity and safeguard the privacy of individuals. So we make sure that we’re doing ethical research. But, you know, ultimately, I think with a lot of the exciting sources of GIS and geotag data, combined with demographic information from the census and other information that we’re getting from social media, I certainly think it would be possible for us to track movements. And as we mine these additional sources of data, you know, we’re going to be using that to address these social and economic questions. But we wouldn’t be on the same scale as if you see in the media, they’re talking about Apple and Google introducing a new means for contact tracing because those data packets are encrypted and handled with their own closed APIs. I don’t think that’s something that we would be able to gain access to. But there are certainly other sources of data that we can use to glean insights into these larger issues associated with the pandemic.

Stuart Shapiro
Great. Let me wrap up by asking you what other goals do you and the RUCI Lab have regarding the analysis of big data to improve life in cities?

Gavin Rozzi
Yes, that’s a great question. So other goals that I have personally, would include investigating the use of emerging technologies to sift through some of these larger datasets. There have been some very exciting developments in the last few years with technologies like web scraping, machine learning, other sources of cloud platforms for warehousing, a lot of this data. I would like to see how we could apply some of these emerging technologies. Another thing I know that we’re going to be doing with the lab is we’re going to be working with drones, and drones are something that I’ve personally found to be really a transformative technology because we can use them to capture very high-resolution imagery, and then overlay that imagery with geographic data, other sources of data. And we can certainly use drones to not only see what’s going on in cities, but also, you know, now that we’re really getting strong advances in machine learning and image processing, we can maybe process that imagery further. And take it from just a static image to, you know, now moving the needle to recognize patterns, recognizing information contained within those images that, you know, 10 years ago would have just been a static image and not really have any other data associated with it.

So I think, you know, the technology is important, but as much as the technology matters, so do people. What we’re doing at the end of the day, is we want to be able to improve the quality of urban living, improve people’s lives. And we’re using technology as a tool to do that. So I think what’s really going to be important as the lab moves forward is the faculty that we’re working with, the outside collaborators we’re working with, either within Rutgers or internationally. And by combining all of that together, the big picture of that allows us to use technology to improve outcomes and give our decision-makers more accurate information, so that we can get into this mindset of data-driven decision making. So that’s where I see the lab going.

Stuart Shapiro
That’s great. Well, Gavin, welcome remotely to the Bloustein School. And hopefully, when we go back physical, we can have some of these follow-on conversations in person and maybe have you back on the podcast, because obviously there’s a lot going on, and a lot to talk about. Thanks for joining us today.

Gavin Rozzi
Thank you for having me, Stuart. I look forward to continuing the conversation.

Stuart Shapiro
A big thank you to Gavin Rozzi and to our production team Tamara Swedberg, Amy Cobb and Karyn Olsen. We’ll be back next week with another talk, with another expert from the Bloustein School. Thanks for listening.

Recent Episodes