Abstract: This talk brings geo-spatial aspect (among other key ingredients) into a real-world application domain, namely the assessment of the impact of social justice campaigns. Besides telling a story, the goal with such campaigns is to motivate change in the knowledge and/ or behavior of individuals and communities. I will present on a project were we are trying to answer the following questions: How can we know if a campaign has had what impact, and how early in the life cycle of a project can we answer this question? The need for the reliable, efficient and systematic evaluation of the impact of media has been repeatedly pointed out by funders, practitioners and researchers (similar efforts are currently happing in science). I will report on our findings and lessons learned from developing, applying and evaluating a theoretically-grounded, computational methodology and technology that facilitate the assessment of media impact in a scalable and comprehensive fashion. We leverage a combination of text mining and network analysis methods to generate socio-semantic networks that are contextualized in space and time. I will dig deeper into a specific detail of this project, namely the identification of various types of locations from unstructured, natural language text data via supervised machine learning techniques.
Bio: Jana Diesner is an Assistant Professor at the iSchool (a.k.a. Graduate School of Library and Information Science) at the University of Illinois, Urbana-Champaign (UIUC), and an affiliate at the Department of Computer Science (CS). Jana’s work is at the nexus of social network analysis, natural language processing and machine learning. With her team, Jana is developing and advancing computational methods and technologies that help people to measure and understand the interplay and co-evolution of information and socio-technical networks. For example, they have been building ConText, a publicly available tool that supports users in extracting network data from text corpora and considering the content of information for network analysis. She brings these computational solutions into various application context, currently mainly in the domains of impact assessment of media and medical informatics. Jana got her PhD from Carnegie Mellon University, School of Computer Science. For more information about her work see http://people.lis.illinois.edu/~jdiesner/.
About the CyberGIS Center Brown Bag Series: The CyberGIS Center Brown Bag Series is a bi-weekly series that brings together researchers from across the University of Illinois, Urbana - Champaign campus to speak about topics that are of intest to the local GIS community.
For more information about the CyberGIS Center visit http://cybergis.illinois.edu/