Euirim Choi

Scoop on how Facebook Inc.’s Instagram offered financial incentives to TikTok users with millions of followers to persuade them to use its competing service, Instagram Reels. The news arrives as TikTok faces an existential threat from the Trump administration and tries to fend off competitors looking to pounce on its rapidly growing userbase.

Notes: The story led the front page of The Wall Street Journal in print and online as the main story. It was aggregated by numerous other media outlets. The next day, TikTok announced that it was growing its creator fund, which appears designed to keep popular TikTok users from exiting the platform, from $200 million to $1 billion.

Leaked emails allegedly show that University administrators favored donors’ children for internship funding even as students in need were turned away in order to strengthen ties between UChicago and its donors.

My co-author and I were invited on WBEZ to discuss our reporting. The story was also featured on Axios, which called it “outstanding.”

Documents reveal that UChicago trustee Steve Stevanovich is late on over $12 million in pledges to the University, and that by overselling his aviation company he may have committed securities fraud.

Notes: Our investigation was profiled by Felix Salmon from Axios, who called it “fantastic.” Our article was cited in federal bankruptcy court.

The almost 5,000-word investigative piece required reading over 1,000 pages of documents. I also coded a special site to host the article and designed the featured graphic.

Sensitive documents obtained by The Maroon last summer and a new lawsuit reveal a strained relationship between the University of Chicago and the Pearson family, calling the future of the Pearson Institute, which was supposed to change the world, into question.

Notes: I broke this story, catching Chicago media outlets by surprise, possibly because the lawsuit was filed in a district court in Oklahoma—not in Illinois. The article was cited on Page A2 of The Wall Street Journal. The Chicago Sun-Times and the Chicago Tribune ran their own reporting on the story on their respective front pages.

I was later subpoenaed by the Pearson family for documents used to produce this article; the subpoena made national headlines and was even covered in the national POLITICO Playbook.

The article is the most-read Maroon article in the past two years. The Illinois Collegiate Press Association said that this was 2018’s best news story in Illinois at the university level. The article won in the large school (4,000+ students) non-dailies category.

Booth professor Luigi Zingales, who is interested in populism, confirmed to The Maroon that Bannon accepted his invitation.

Note: I waited for two hours outside Professor Zingales’ office to confirm a tip that we received. We then published a story before every outlet, which would later be confirmed by the Chicago Sun-Times and the Chicago Tribune. The Washington Post wrote about the invitation as well.

A Maroon investigation found that weaknesses in the University’s network could have allowed hackers to steal sensitive information from networked printers, as well as access and, in some cases, control cameras and other connected devices.

Notes: The Illinois Collegiate Press Association said that this was 2017’s best news story in Illinois at the university level. The article won in the large school (4,000+ students) non-dailies category.

Private administration documents shed light on strategic budget deficits and deliberations on how to control spending.

Notes: This article was part of a series that was deemed by the Illinois Collegiate Press Association as being 2017’s best in-depth reporting series at the university level. The series won in the large school (4,000+ students) non-dailies category.

Tracking the number of Twitter mentions in real-time of each major candidate in the 2020 U.S. Presidential Election. I used ArchieML, a library by The New York Times, to organize page content.

In this paper, we develop a cost-benefit model that uses housing and demographic data from the U.S. Census, solar irradiance and electricity price data from the National Renewable Energy Laboratory, and roof segmentation data from Google’s Project Sunroof to study cost-effectiveness by zip code of purchasing one particular model of a rooftop solar installation, the Tesla solar roof and Powerwall.

We find that zip codes in the arid, sunny Southwest have the highest cost-effectiveness, while Appalachia and western Washington have the lowest cost-effectiveness, due to low electricity prices and low solar irradiance, respectively. We also discuss the impact of different federal and state subsidy models on cost-effectiveness.

Along with assisting in the data preprocessing and analysis, I made and designed the maps displaying our results using QGIS and Adobe Illustrator.

An online buy-sell platform used by over 10,000 people in Hyde Park and its surrounding neighborhoods in Chicago. I built it using React, Django, Docker, and Amazon AWS.

Every year millions of people around the world read The Chicago Maroon’s journalism online. I rebuilt the website from the ground up using Django as well as JavaScript and moved the site from being hosted locally to being supported by Amazon AWS.

The Maroon received an award from the Illinois College Press Association for being the best university newspaper website in Illinois in 2018. It won in the large school (4,000+ students) non-dailies category.

Ben Weinshel, Miranda Wei, Mainack Mondal, Euirim Choi, Shawn Shan, Claire Dolin, Michelle L. Mazurek, Blase Ur

Internet companies track users’ online activity to make inferences about their interests, which are then used to target ads and personalize their web experience. Prior work has shown that existing privacy-protective tools give users only a limited understanding and incomplete picture of online tracking. We present Tracking Transparency, a privacy-preserving browser extension that visualizes examples of long-term, longitudinal information that third-party trackers could have inferred from users’ browsing. The extension uses a client-side topic modeling algorithm to categorize pages that users visit and combines this with data about the web trackers encountered over time to create these visualizations. We conduct a longitudinal field study in which 425 participants use one of six variants of our extension for a week. We find that, after using the extension, participants have more accurate perceptions of the extent of tracking and also intend to take privacy-protecting actions.

Claire Dolin, Ben Weinshel, Shawn Shan, Chang Min Hahn, Euirim Choi, Michelle L. Mazurek, Blase Ur

Much of what a user sees browsing the internet, from ads to search results, is targeted or personalized by algorithms that have made inferences about that user. Prior work has documented that users find such targeting simultaneously useful and creepy. We begin unpacking these conflicted feelings through two online studies. In the first study, 306 participants saw one of ten explanations for why they received an ad, reflecting prevalent methods of targeting based on demographics, interests, and other factors. The type of interest-based targeting described in the explanation affected participants’ comfort with the targeting and perceptions of its usefulness. We conducted a follow-up study in which 237 participants saw ten interests companies might infer. Both the sensitivity of the interest category and participants’ actual interest in that topic significantly impacted their attitudes toward inferencing. Our results inform the design of transparency tools.