The Large Scale Video Analytics (LSVA) project’s primary aim is to construct an environment whereby researchers may use the Gordon Supercomputer to transform video data, search through large video archives for emergent and large-scale patterns, and annotate the data according to their findings. We propose to do this by modifying the NCSA Medici web based content repository and adding support for content based video retrieval.


Principal Investigator: Dr. Virginia Kuhn
Institution: University of Southern California

Real Stories of Bad Kids — A Historical Big Data Analysis to Disclose the Social Construction of Juvenile Delinquency

This project deals with studying the social construction of juvenile delinquency in the United States using data analysis of scanned newspapers from the 19th and 20th century. Social construction is a theoretical position that social reality is created through the humans’ definition and interaction. As one type of social reality, juvenile delinquency is perceived as part of social problems, deeply contextualized and socially constructed in American society. The social construction of juvenile delinquency has started far earlier than the first juvenile court in 1899 in the U.S. Scholars have tried traditional historical analysis to explore the timeline of the social construction of juvenile delinquency in the past, but it is inefficient to examine hundred years of the document using traditional paper-pencil documenting method. This project will combine image and linguistic analyses, and big data tools to analyze hundreds of years of scanned newspaper images and show a clear development of social construction of juvenile delinquency in the American society. The project will explore the following questions: how the media (one type of social construction) start depicting certain types of juvenile behavior as delinquency, how they describe those behaviors; who are those juveniles (age, race, gender, family background, community background), how other social institutions treat those juveniles in those stories; how the depiction of juvenile delinquency has changed during the past 100 years; whether the analysis results support social construction perspective in terms of juvenile delinquency. This project will explore its research objectives in a collection of digitized print texts. During the startup phase of this project, we will use data from an archive of newspapers (1853-1921) from http://chroniclingamerica.loc.gov/newspapers/. We are investigating sources for newspapers that are newer than this archive provides, but due to copyright restrictions, etc. we will focus on this collection during the startup phase. The techniques and tools we create during the startup phase should also work on newer sources when we gain access to them.
Principal Investigator: Dr. Yu Zhang
Institution: The State University of New York at Brockport