Eslami, Shang win Google Fellowships
By Enrique Rodarte, Graduate College
Launched in 2009, The Google US/Canada PhD Fellowship Program recognizes outstanding graduate students doing exceptional work in Computer Science and related disciplines, such as Computational Neuroscience, Machine Learning, and Market Algorithms.
Eslami’s research interests include social computing, data mining, and human-computing. In particular, her research explores the increasing prevalence of algorithms in sociotechnical systems and how users understand and relate to them.
"Algorithms," she explains, "curate everyday online content by prioritizing, classifying, associating, and filtering information. Through this curation, algorithms exert power to shape user experience and even the evolution of the system as a whole. Algorithms are, however, usually hidden in black boxes to protect intellectual property and to hide details from users to make their interactions with the system effortless. This black box nature can prevent users from understanding the details of algorithms functionality or even their existence, but users perceived knowledge about an algorithm can still affect their behavior. For instance, when believing that the Facebook News Feed curation algorithm ranks posts with commercial keywords higher, some teenagers added product names to their posts in an attempt to manipulate the algorithm and increase their posts’ visibility."
Shang’s research looks at ways to help structure the massive amount of information we create. "My research focuses on turning unstructured text data into structured knowledge with very little human effort. The majority of the massive amount of data in the real world – including news, social media, scientific papers, government documents, business contracts, and web content – are unstructured or loosely structured text. The sheer size of such data and the fast pace of new data generation make manual labeling unscalable and infeasible. Automated structure discovery and construction of knowledge graphs from massive text corpora have become an active research area in the fields of natural language processing, machine learning, and data mining, but existing approaches rely heavily on human annotation and curation.”
The Google US/Canada PhD Fellowship offers two to three years of support, including an annual stipend of $34,000, coverage of tuition and fees, and a Google research mentor. Applicants must be nominated by their department. The Graduate College then facilitates a faculty review panel, which selects nominees to forward to Google.
The award is very selective. Typically only 12-15 students are selected for the awards in any given year. Previous University of Illinois recipients include: Ali Farhadi and Eric Gilbert (2009), Maryam Karimzadehgan (2010), Hongning Wang and Ankit Singla (2012), Muhammad Naveed (2015), and Xiang Ren (2016).
Ken Vickery, director of External Fellowships in the Graduate College at the University of Illinois at Urbana-Champaign, shares his enthusiasm: “With these awards, Illinois students continue to demonstrate the remarkable vision and originality of their research. Google fellowships go only to the top students worldwide in computer science and related fields, and having two Illinois students selected this year speaks to the exemplary promise of these students and as well as the excellence of our graduate training programs.”
Information about the Google US/Canada PhD Fellowship program can be found at: https://research.google.com/research-outreach.html#/research-outreach/graduate-fellowships