SHANTI at the University of Virginia

News / Latest Posts

Computer Science Guest Speaker: Ahmed Abbasi


Ahmed Abbasi
McIntire School of Commerce

“Text Analytics to Support Sense-making in Social Media: A Language-Action Perspective”

Friday, October 28, 2016

3:30 pm (short reception before talk at 3:00)

Rice Hall Auditorium

Host: Hongning Wang (hw5x)

Abstract: This presentation summarizes the results from a five-year text analytics project examining social media sense-making capabilities in various industries, including telecommunications, health, and security. Despite their various benefits, social media technologies present two important challenges for sense-making. First, online discourse is plagued by incoherent, intertwined conversations that are often difficult to comprehend. Moreover, existing text analytics tools mostly focus on the semantic dimension of language, as opposed to actions and intentions. I use real-world examples to illustrate how these challenges inhibit our ability to perform many basic social media analytics tasks such as identifying important participants, issues, and ideas. The Language-Action Perspective (LAP) emphasizes pragmatics; considering conversations, actions, and context. In order to address the two aforementioned challenges, we developed a LAP-based text analytics framework to support sense-making in online discourse. I present evaluation results from multiple social media channels and industries, including an extended field experiment utilizing a large cloud-based system developed based on the framework. The results have important implications for online sense-making, social media analytics, and how we think about text.

Biography: Ahmed Abbasi is Murray Research Professor and associate professor of Information Technology in the McIntire School of Commerce at the University of Virginia. He is Director of the Center for Business Analytics and coordinator for the Enterprise IT Management module of the MS in MIT executive degree program. Ahmed is also a member of the Predictive Analytics Lab. Ahmed received his Ph.D. in Information Systems from the University of Arizona, where he also worked as a project-lead on multi-million dollar “big data” initiatives in the Artificial Intelligence Lab. He attained an M.B.A. and B.S. in Information Technology from Virginia Tech.

DH@UVa Conference

DH@UVA Conference

Friday, October 14 – Saturday, October 15
Auditorium, Harrison Institute & Small Special Collections Library

Please join us for two days of conversation and insights with the digital humanities community at UVa.

On October 14 and 15, the digital humanities community at UVa will come together for a multiplicity of conversations about what is happening in DH here now. As our community of practice grows and evolves, we are taking a moment to reflect and exchange ideas about what can happen in the future. Join us for lighting talks, roundtables, and ongoing public dialogue. Invited speakers include Lauren Klein, Tanya Clement, and Chris Johanson.

To learn more and register, please visit DH@UVa 2016 is free and open to the public. It is initiated and sponsored by Ron Hutchins, Vice President for Information Technology, Archie Holmes, Associate Provost for Academic Affairs, Francesca Fiorani, Associate Dean for the Arts and Humanities, and John Unsworth, Dean of Libraries and University Librarian. It is organized by Alison Booth, Academic Director of the Scholars’ Lab; David Germano, Director of SHANTI; and Worthy Martin, Director of IATH, with the support of Judy Thomas, Senior Director of Academic Engagement, UVa Library.


ARCS: Registration for Fall 2016 Workshops Now Open

Registration for Fall 2016 Workshops Now Open

ARCS workshops are absolutely FREE and open to any member of the UVA community with an interest in learning more about advanced computing. Workshops last for 2 hours and are normally offered during the fall semester.

Fall 2016 workshops:

Introduction to NumPy, SciPy, and Matplotlib (Instructor: K. Holcomb)
Effective Use Of HPC Systems (Instructor: K. Holcomb)
Image Processing with ImageJ (Instructor: K. Siller)
Intro to Bash Scripting (Instructor: K. Siller)
Optimizing R Code (Instructor: J. Huband)
Introduction to Parallel R (Instructor: J. Huband)
Parallel Mathematica (Instructor: Ed Hall)
Parallel Matlab (Instructor: Ed Hall)​ ​


DH Speaker Series: Toniesha Taylor


Digital Humanities Speaker Series presents:

Toniesha Taylor
“We Speak, We Make, We Tinker:
Afrofuturism as Applied Digital Humanities”
Friday, September 23
10:00 am · Alderman, Rm 421




Afrofuturism is the manifestation of digital humanities. The goals of the thinkers, makers, creatives, and speakers involved in both digital humanities and afrofuturism are so much the same—yet—impossibly different. Afrofuturists focus on the future past were Black peoples are people—more than flesh—they are humanities. They have powers and abilities so present and ancient they are as unreadable script as the 3D maker spaces of digital humanist are to the CD ROMs of your average humanist. We are the hidden code. Don’t look for the gatekeepers to lock us out, our imaginations know no gates. We don’t wait for the dust to settle on old debates. We’ve already made new ideas and new dust.

Digital Humanities is about making a past future where all of human knowledge and creation is understood to benefit human future past. To achieve this goal it can’t duplicate the failure of humanities by excluding human women, Black humans, Gay humans, Transgender humans, Asian humans, Queer humans, differently able humans. Digital humanities must first recognize humanness before it can code and create visualizations of a past future or future past. Afrofuturism can help.

This talk focuses on the ways in which Afrofuturism and Digital Humanities can come together to bring their inherent creative theories, methods and applications together to engage in an equitable discourse to changes the future of the humanities. I argue that born out of these collaborations could be new approaches which will engage knowledge productions in the areas of the humanities often relegated to “area studies” and recenter those contributions within their equitable portion of human knowledge.

Toniesha Taylor is an Associate Professor of Communication in the Department Languages and Communication at Prairie View A & M University. She engages with discussions on womanist rhetoric as method and theory; practical social justice pedagogy for faculty and students; critical engagement in popular cultural critique; digital humanities methods implications for activist recovery projects; African American women’s sermons and conversion discourses both historic and contemporary. Her research projects include the Prairie View Women’s Oral History Project, designed to collect, preserve, curate and display the oral histories of women at Prairie View A&M University, and “White Violence, Black Resistance,” with Dr. Amy Earhart, which seeks to digitize a broad set of primary documents related to interactions of race and power. Both projects can be found at

9/30/16 — SEI Colloquium Series: Dan Cosley

The SEI Colloquium Series presents:

Speaker: Dan Cosley, Ph.D.
Date: 30 September 2016
Time: 2:00 – 3:15pm
Location: 120 Olsson Hall

Note: Please join the speaker for light refreshments in Olsson Hall, Room 104 at 1:30 p.m.


The notion that ideas, products, and innovations spread “virally” through information networks resonates in both popular and scientific accounts of how things become popular. These accounts often focus on the role of others’ influence, using models that emphasize structural roles in networks (Gladwell’s “mavens, connectors, and salesmen”; Keller and Berry’s “influentials”) or make analogies to the spread of diseases (threshold and cascade models rooted in epidemiology). In this talk I will argue that such models are poor fits because they don’t account for the agency of those on the receiving end: paying attention to an idea is not a disease that you catch but a choice that you make. Based on work with Dr. Amit Sharma that looks at how people’s choices are affected by social explanations of recommendations (e.g, “Thomas Jefferson and 9 of your friends like this”) and feeds of friends’ activity (as in, Flickr, and other sites that show us our friends’ actions on items), we conclude that personal preferences dominate decision making in a way that models of information diffusion don’t, but should (and increasingly could), account for. Along the way we’ll look at considerations for designing more effective social explanations, methods that try to tease out social influence from underlying personal preferences, and ways theory might help us get a more nuanced handle on the broad notion of “influence” by focusing our attention on specific mechanisms.

About the Speaker:

Dan Cosley is an associate professor in information science at Cornell University, currently on rotation as a program officer in NSF’s Cyber-Human Systems program, who does research broadly around human-computer interaction and social media. His high-level research goal is to build and study systems that leverage people’s online behavior to improve both individual and social outcomes. He has done such work in a number of domains, including modeling information dissemination in social media, supporting civic participation online, using social media content to support reminiscence and reflection, helping people contribute to public goods such as Wikipedia, and improving recommender systems. Much of the work has been supported by NSF and is rooted in the PhD in computer science he received in 2006 under the guidance of advisors John Riedl and Loren Terveen from the University of Minnesota.