Projects

We study how operation and platform strategy affect product- and firm-level innovativeness in the smartphone industry. Our study is based on the technological evolution of smartphones including physical characteristics, performance characteristics, and features over the past decade. Our longitudinal analysis of a dataset consisting of 1,171 smartphones by 79 device manufacturers and 12 different platforms highlights the significant transformational changes that have occurred. We propose an index to measure innovativeness of a smartphone upon which we test our main hypotheses.
Followers are Twitter's most basic currency. Building an audience of followers can create access to a network of social ties, resources, and influence. Yet, little is understood about how to grow such an audience. We examine multiple factors that affect tie formation and dissolution over time on the social media service Twitter. We collected behavioral, content, and network data approximately every three months for fifteen months and compared the relative contributions of 22 variables from each perspective for predicting link formations in online social networks.
Photos are becoming prominent means of communication online. Despite photos' pervasive presence in social media and online world, we know little about how people interact and engage with their content. Understanding how photo content might signify engagement, can impact both science and design, influencing production and distribution. One common type of photo content that is shared on social media, is the photos of people. From studies of offline behavior, we know that human faces are powerful channels of non-verbal communication. In this paper, we study this behavioral phenomena online.
The Power Puppet project combines the creation of a mechanical puppet with simple circuit building. Its goal is to teach middle school students basic circuit building in the setting of a puppet building workshop. As students build their puppets, including control mechanisms (like rods and strings) and expressive elements (like joints and materials), they also create basic circuits that operate in combination with the puppet that houses them.
This is the first is a series of experiments in locative narrative. In this version, Projet is entirely screen-based. Later applications in this series will use Argon to create true location-based narratives. Projet itself can be experienced on a laptop or desktop screen, but there is greater intimacy and effect when experienced on a tablet or phone. A woman finds herself in a village in the French countryside, reflects on her present life, and recalls her past. Projet reduces the viewer's interaction to the barest minimum.
Chatbots are one of the newest media in consumer-facing technologies. A chatbot can be defined broadly as any service that users interact with primarily through text. More and more companies and organizations are choosing to communicate with their users through bots. There are a couple reasons for this: one, the barrier to entry is relatively low since users don't have to download a new app or learn a new UI; and, two, since chatbots communicate primarily through text, a bot affords an organization the opportunity to add personality to their brand.
We combine craft and performance art to teach early middle school students basic prototyping skills. We develop informal STEM workshops for puppetry that combine narrative framing, craft-inspired building, and performance. This key approach combines craft, art, and basic hardware prototyping to attract new audiences to STEM. It is a collaboration between Georgia Tech and at the Center for Puppetry Arts funded by the NSF.
Designing a system to provide real-time feedback on Nurse's compliance in the use of personal protective equipment when entering a patient's room in the intensive care unit.
Researchers in ubiquitous computing fields have assumed that smartphone owners consistently use, carry, or have access to their phones throughout the day. Current applications in fields such as mHealth believe that the smartphone is a promising and reliable sensing platform to understand user behaviors and environment. However, as wearable technology (e.g., smartwatches) has become more widely adopted, the changed technology landscape has impacted users' proximity to their smartphone. By introducing new research techniques, we investigate the current users' proximity to their smartphone.
Psi and Delta is a collaborative classroom videogame that enables students to experience the world of the very small, together.
This ongoing study investigates the effect that proximate criminal activity has on emotional expression in social media. Proximity to crime and well constant fear of crime can have great negative psychological effects on individuals. Social media currently being one of the most popular means of publicly expressing personal opinions and emotions, we expect to find an effect of temporal and spatial proximity to crime on social media mood expression and other patters of online communication.
The Pueblo Connect project is focused on investigating methods for measuring cellular connectivity in rural areas and, more broadly, increasing access to the internet in rural and indigenous communities.
PUNGA (Provenance-supported Undirected Node Graph Analytics) is a tool for intelligence analysts. PUNGA assists analysts in making sense of a large textual-based dataset by supporting data processing (Named Entity Recognition), data cleaning, data analysis, and analytic provencance. PUNGA provides users the ability to combine, format, clean the data as per their convenience before and during analysis with the Entity View. PUNGA also facilitates user interaction with the data sets in a number of linked views.
Visual analytics (VA) systems with semantic interaction help users craft machine learning (ML) based solutions in various domains such as bio-informatics, finance, sports, etc. However current semantic interaction based approaches are data and task-specific which might not generalize across different problem scenarios. In this project, we describe a novel technique of abstracting user intents and goals in the form of an interactive objective function which can guide any auto-ML based model optimizer (such as Hyperopt, Sigopt, etc.) to construct classification models catering to the expectation

Pages