Faces in Places: Real-time Diversity Analytics Using Social Media (May 2016 – August 2017)
The objective of this work is to determine whether a social media-based computational framework can be employed/deployed to obtain diversity, and segregation scores at very fine (i.e. individual relationships) resolution. Specifically, we focus on Instagram photos of multiple people interacting, and employ automatic methods for race, age, and gender estimation to quantify mixing in such photos. Results obtained motivate the use of social media photos to complement census data to develop cheaper, faster, mechanisms for studying diversity and applying them in social, economic, political, and urban planning contexts. Applications of this have been extended to understanding whether a similar framework can be employed to obtain smile and diversity scores at very fine, individual relationship resolution, and study their associations via Twitter and Instagram.
E-petition Information Diffusion in Online Social Networks (Jan 2016 – Feb 2017)
E-petitioning, a genre of technology-based collective action tools, makes it possible for members of the public to address government decision makers directly with their requests for action. In this work, we use time series analysis to explore the effects of Twitter and other forms of online media on the accumulation of signatures in e-petitioning. We explore the case of “Bring Back Our Girls,” a Change.org petition initiated in spring 2014 following the abduction of 276 female students from a school in Chibok, Nigeria. The petition targeted government leaders around the world. We found evidence that tweeting and certain forms of online media are related to the likelihood of individuals signing an e-petition, providing evidence of a hybrid media system in which diverse forms of online media behave with diverse logics and impacts in their effects on e-petitions.
Twitter Popularity Diffusion of Presidential Candidates through Detection of Twitter Bots (Aug 2016 – Jan 2017)
The objective of this work was to detect popularity diffusion patterns on Twitter in the 2016 US Presidential Election and consider the role, implications and effects of Twitter bots.
Reeb Graph Based Validation of Statistical Predictive Models for Spread of AIDS (Nov 2014 – Dec 2015)
This work presents a method to quantify the differences in data using a newly defined distance metric between graphs that is used to study, analyze and validate a statistical predictive model. The proposed method constructs smoothed Reeb graphs of a given graph, which is then used to compare different outcomes of the statistical model. The method to was intended to study the given AIDS data by analyzing the output of the models in order to identify how relationships among people in a community affects their susceptibility to AIDS. In order to do this, we have applied ideas from computational topology techniques to be able to do statistics on graphs.