Skip to main content

Computing student earns conference scholarship from Google


Lisa Tsosie
September 11, 2013

Her work on research projects, including developing a tool to help parents identify bullying of adolescents on Facebook, has earned Arizona State University student Lisa Tsosie a Google Women of Color Scholarship to attend the 2013 Grace Hopper Celebration of Women in Computing, Oct. 2-5 in Minneapolis. Tsosie was selected to receive one of the highly competitive awards from more than 900 applicants.

The annual Grace Hopper Celebration conferences are designed to help advance the careers and research of women in computing. Leaders in industry, academia and government make myriad presentations, while special sessions focus on the role of women in today’s technology fields, including computer science, information technology, research and engineering. This year’s featured speakers include Sheryl Sandberg from Facebook and Ana Pinczuk from Cisco Systems.

Tsosie is a junior majoring in applied computing through ASU’s New College of Interdisciplinary Arts and Sciences, the core college on the West campus. She was encouraged to apply for the scholarship by Yasin Silva, assistant professor in New College’s School of Mathematical and Natural Sciences, who teaches courses in the applied computing program.

Silva and Tsosie worked together to produce the abstract “Facebully: Towards the Identification of Cyberbullying in Facebook,” which was selected for presentation at the Grace Hopper Celebration. But Tsosie didn’t expect to be able to attend.

“I didn't get my hopes up because I didn't think I'd be able to afford the trip, so I assumed Dr. Silva would be traveling and presenting our results,” she said. “I was definitely proud and encouraged that the committee wanted me to present a poster on our research and had rated our abstract so positively.”

Receipt of the Google scholarship enabled Tsosie to attend the conference.

“My understanding is that the conference is all about reaching out to women in technology, and I feel privileged to be a part of this celebration,” said Tsosie, whose ethnic heritage is Navajo and Romanian. “I hope that I can expand my female, professional network from participating and that by sharing my experience I might draw other women into this field as well.”

Tsosie describes cyberbullying as the most common online risk for adolescents. She and Silva have been working on the “Facebully” project since her sophomore year.

“Our main goal is to design and implement a model to identify cyberbullying in Facebook, building on previous research findings in the areas of traditional and cyberbullying with adolescents,” Silva said.

“The model uses social media data, including content of posts, changes of schools or neighborhoods, and variations in social interaction patterns, such as a sudden significant increase in the number of posts, to compute a Bullying Rank function,” he said. “A high value of this function indicates high probability that an individual is experiencing cyberbullying. Once the application is ready for deployment, it can be used by parents to monitor their children via their social network and give them knowledge of whether their child is a victim of online aggression.”

Added Tsosie, “While previous research has identified the different measures of cyberbullying through venues more frequently utilized to mediate online aggression, like mobile devices and chat rooms, there has been a limited amount of research in the area of automatic identification of cyberbullying in social media sites, particularly in Facebook.”

Tsosie also is working with Silva on the study, evaluation and comparison of complex distributed algorithms to identify pairs of similar objects in very large datasets.

“A Similarity Join is an operation that identifies all the pairs of similar objects in a dataset and is extensively used in multiple application scenarios,” she said. “Many of these scenarios require processing massive amounts of data, but with Cloud systems and MapReduce, we can process this data in a highly scalable and distributed fashion. There have been several solutions proposed and published, but not a clear understanding of how these algorithms compare to each other. Our research seeks to identify which approach is best for a specific set of requirements.”

In addition to presenting the “Facebully” research poster at the Grace Hopper Celebration conference next month, Tsosie has co-authored what Silva describes as two high-quality publications and is also a co-author in a third paper currently under submission. Portions of Tsosie’s research have been conducted as part of projects funded by two grants from the Western Alliance to Expand Student Opportunities.

“Lisa’s outstanding record on research is rare for an undergraduate student,” Silva said. “She is both a highly skilled and very hardworking young researcher.”

In addition to her pursuit of a bachelor's in applied computing with a database concentration, Tsosie also is pursuing a writing certificate through New College’s School of Humanities, Arts and Cultural Studies.

Attending the Grace Hopper Celebration conference should be highly beneficial for Tsosie on multiple levels, according to Silva. “She will be able to present the results of her research work, network with other women who have similar technical interests and participate in special sessions focused on exploring graduate school and professional opportunities,” he said.