Eyes darting or maintaining a steady gaze straight ahead. Heartbeat racing, or maintaining a slow, even rhythm. If we encounter these phenomena in another, how do we respond – not just effectively, but physiologically? Eye movements and heartbeats are among the most intuitively meaningful physiological characteristics that humans observe in one another. Without necessarily consciously realizing it, we often respond empathetically.
This map visualizes demographic indicators of gentrification in neighborhoods (defined by census tracts) along the current and proposed path of the Atlanta Beltline, an "urban redevelopment" project under construction along a loop of disused railroad tracks that circumvent the city, stitching together some of its most historic neighborhoods.
As time has gone by, research has pointed out that there remains skepticism in the community when it comes to using such tools that are focused around automating parts of the data discovery process. As my research as well as previous research has shown, practitioners prefer using traditional tools over newer tools, even if it involves more manual effort and time. The workflows that many data scientists follow today, do not completely utilize the new capabilities that are available to them through ML/AI.
Many electronic devices, from desktop computers to mobile phones to DVD players, can be thought of as a menu of functions. These functions can be accessible to a blind user if the menus are spoken aloud. However, this is extremely inefficient, so we have been enhancing auditory menus with sophisticated text-to-speech, spearcons, spindex, and other audio extensions. These can also be applied in many different languages and research is ongoing to look at more language applications, including tonal types.
Networking and peer inspiration from alumni of your program/school is important when making decisions about the next steps in your career. However, schools lose touch with alumni once they graduate and find it difficult to keep a track of where they are. Networking platforms such as Linkedin are helpful but do not provide a big picture of your alumni network. AlmaBase is a Linkedin extension, that shows a visualization of career trajectories of alumni from your program, for you to find the "right" alumni to network with and get inspired.
This project, the winner of the 2020 EmPOWER Air Data Challenge from the US EPA creates an online tool for exploring relationships between coal power plant emissions, wind currents, and local health impacts.
Leveraging publicly available reddit API in order to extract data and then perform relevant machine learning analysis to resulting visual interface tool in order to analyze the results.
Atlanta has the reputation of being a "city in a forest" with a large and varied tree population that provides shade for its residents, a habitat for wildlife, consumes carbon dioxide from the atmosphere, produces life-giving oxygen, in addition to many other benefits. In keeping with its context and commitments to environmental awareness and conservancy, the Georgia Tech campus contains hundreds of species of trees that cover the landscape.
The goal of the Atlanta Map Room is to document and reflect upon the connections and disjunctions between civic data and lived experience in the city, through the collaborative creation of large-scale, interpretive maps.
The graphs and figures that are so prevalent in math and science education make those topics largely inaccessible to blind students. We are working on auditory graphs that can represent equations and data to those who cannot see a visual graph. A number of new areas we're starting research on are: looking into teaching astronomy concepts through (like the Solar System) and the teaching and understanding of weather information through a combination of sonification and auditory description.
As a college student, finding your internship or a job post graduation can be a daunting task. As a novice new to industry or even someone with experience, finding a job tends to be a black box experience. Deciding where, when and how to apply is a process that could benefit from improvement. Career services and fellow peers are great data resources for the job search however often this data is either not accessible or its not in a form that is easy to digest. On the flip side, career professionals such as career service office and program administrators constantly work with students to find
Automated safety systems, a first step toward autonomous vehicles, are already available in many commercial vehicles. These are systems such as adaptive cruise control, which has the capability to slow down due to traffic, and automatic lane keeping, which maintains position within a lane without driver intervention. In order to ensure that these systems are properly used by drivers it is essential that they understand and appropriately trust the technology.
Visualizations can help amplify human cognition. In an era where networks are becoming increasingly complex, the desirability of tools to compare and contrast sets, relationships, and reach is significant. Motivated by a practical need articulated by corporate decision makers, this research presents our journey in designing and implementing bicentric diagrams, a novel graph-based set visualization technique. A bicentric diagram enables simultaneous identification of sets, set relationships, and set member reach in integrated ego networks of two focal entities.
Static visualizations have analytic and expressive value. However, many interactive tasks cannot be completed using static visualizations. As datasets grow in size and complexity, static visualizations start losing their analytic and expressive power for interactive data exploration. Despite this limitation of static visualizations, there are still many cases where visualizations are limited to being static (e.g., visualizations on presentation slides or posters). We believe in many of these cases, static visualizations will benefit from allowing users to perform interactive tasks on them.
Campus Tour is an augmented reality experience of Georgia Tech's campus. Once the channel is loaded in Argon, a standards-based Augmented Reality (AR) web browser developed by the Augmented Environments Lab. The tour gives information to users through text, pictures and videos. Stops on the tour are panoramic images.Within the panoramas are points of interests that once clicked reveal more information about their topic. Campus Tour allows users to remotely enjoy the beauty of campus or to learn more about Tech while on campus.
Healthcare big data is being widely touted as a potential resource for curbing costs and improving outcomes. However, numerous challenges remain for leveraging this data to its full potential. In this position paper, we identify the difficulties that characterize clinical data, based on our experiences working with pediatric asthma data from Children's Healthcare of Atlanta.
The Computational Enterprise Science Lab focuses on the design, analysis, and management of complex enterprise systems (e.g. organizations, supply chains, business ecosystems) using information visualization, modeling/simulation, and system science approaches.
This project studies ways to address counterfeiting through supply network designs and policies. Of particular interest is how such designs and policies may produce unintended effects. A large-scale interactive simulation model has been developed as a test-bed for various designs and policies to demonstrate their effect and their interactions. The policy-maker can execute the model with his or her policy and design choices, and the visual interface will show the results.
Designing effective CSCW systems in healthcare requires a careful consideration of the entire enterprise. This study uses computational text analysis and network visualization of topical terms and keywords to map the extant knowledge domain of CSCW in healthcare. The results are framed using a multi-level enterprise model, comprised of people, technology, process, and organization. Emerging trends and prominent patterns are identified. The study contributes to a broader understanding of CSCW research in healthcare and demonstrates the value of adapting an enterprise (as a) system lens.
An Interactive History of Data Visualization, 1786-1900
Data feminism is a way of thinking about data science that is informed by the past several decades of feminist activism and critical thought.
Data Illustrator is a system that helps graphic designers created data-driven visualizations and infographics without the need to do any programming. More specifically, it is a vector editing tool for creating data visualizations and infographics. Graphic designers can use Data Illustrator to craft their own visualizations by repeating and styling shapes with data-driven rules. The tool supports the creation of expressive, flexible, and parametrically defined visualizations without the need to program them.
Debate Slates is a second screen application experience designed to facilitate discussion of theories and future plot developments of long-form narrative television (e.g. Game of Thrones, True Detective and Fringe); Debate Slates also hopes to facilitate discussions on current events, focusing on televised reportage of ISIS and the developing situation in Iraq and Syria.
Design and Social Justice Studio brings an interdisciplinary group of faculty and students together to examine the experiential and participatory dimensions of digital media and their relationship to establishing and supporting democratic forms of social interaction.
Most public sources reporting air quality data either present the value without sufficient supports for deeper exploration of multiple pollutants or are robust data repositories that are too technical to be accessible to non-scientific audiences. Our findings indicate many people have little context for understanding how the Air Quality Index (AQI) is generated or what it measures. In response, we are creating a contextualized, visualization-based platform to support public audiences in exploring air pollution beyond the AQI by displaying contextualized multi-pollutant data.
The Digital Humanities Lab (DH Lab) explores how digital technologies engage questions of humanistic inquiry.
The Covid-19 pandemic has certainly caused a lot of disruption in how universities operate. Proactive campuses use a combination of testing, tracing, and encouraging safe practices to contain infectious spread while still maintaining some semblance of operation. In this video, we will present some work on using WiFi authentication logs to support contact tracing (identifying potential contacts) as well as understanding the impact of policy (e.g., online teaching only) on infectious spread.
Business ecosystems are characterized by large, complex, and global networks of firms, often from many different market segments, all collaborating, partnering, and competing to create and deliver new products and services. Given the rapidly increasing scale, complexity, and rate of change of business ecosystems, as well as economic and competitive pressures, analysts are faced with the formidable task of quickly understanding the fundamental characteristics of these interfirm networks.
Business ecosystems are characterized by large, complex, and global networks of firms, often from many different market segments, all collaborating, partnering, and competing to create and deliver new products and services. Given the rapidly increasing scale, complexity, and rate of change of business ecosystems, as well as economic and competitive pressures, analysts are faced with the formidable task of quickly understanding the fundamental characteristics of these interfirm networks.
How might we create confidence and increase communication to riders of a public transportation system? This project is focusing on creating an interactive display and experience for riders of the Georgia Tech Bus system. The goals of this display is to provide multi-sensory, at-a-glance information about the bus system in relation to the station's current location, while also providing more detailed information on closer inspections for users that wish to make a more informed decisions around using the bus.
In an increasingly global and competitive business landscape, firms must collaborate and partner with other firms to ensure survival, growth, and innovation. Understanding the evolutionary composition of a firm's relationship portfolio and the underlying formation strategy is a difficult task given the multi¬dimensional, temporal nature of the data.
We have created a visualization that presents one week of a person's Facebook messages and notifications. The focus here is to allow someone to quickly catch up with what has been going on in their feed, which messages were "hot", who has been active, etc. The tool leverages a number of different visualization techniques and can benefit from a very large display.
Analysts routinely encounter large collections of text documents with many accompanying numeric or categorical data fields. For example, consider a collection of wine reviews where the main text part of the document is the actual review narrative, but the accompanying fields are data such as the wine's variety, color, age, rating, producer, region, reviewer, and so on. In this project, we are exploring techniques to allow analysts to investigate the document collection by "slicing" it along the different attributes affiliated with each document (we call them "facets").
A textual visualization that analyzes and presents beer review data to allow users to find a beer that meets their tastes.
Researching how women in Delhi deal with menstrual health outside of their home with the goal of designing an application to aid them in categorizing and finding bathrooms and other safe spaces to meet their needs.
Graph-Level Operations (GLOs) are a holistic vocabulary of encapsulated manipulations of graph visualization elements. GLOs allow analysts to explore their network data in new and interesting ways, freeing them from being restricted to predefined graph visualization techniques. GLOs also provide software engineers with an alternative, extensible means of writing extensible graph visualization applications. Finally, GLOs provide an elegant method for generating animated transitions between graph visualization techniques.
Many real-world datasets are large, multivariate, and relational in nature and relevant associated decisions frequently require a simultaneous consideration of both attributes and connections. Existing visualization systems and approaches often make an explicit trade-off between either affording rich exploration of individual data units or exploration of the underlying network structure. We demonstrate through a system called Graphicle how network structure can be layered on top of unit visualization techniques to create new opportunities for visual exploration of large multivariate networks.
Network visualizations, often in the form of node-link diagrams, are an effective means to understand relationships between entities, discover entities with interesting characteristics, and to identify clusters. While several existing tools allow users to visualize pre-defined networks, creating these networks from raw data remains a challenging task, often requiring users to program custom scripts or write complex SQL commands. Some existing tools also allow users to both visualize and model networks.
The annual Clough Commons Art Crawl serves as a unique opportunity for Georgia Tech students to close their books, catch their breath, and enjoy the therapeutic effects of art. The blank walls of the Clough Commons will once again be transformed into a make-shift gallery, all centered around the artistic work of Georgia Tech students.
This glassware is designed for the Georgia Tech campus community and visitors. It uses your location information to help you know what buildings are nearby as well as find the nearest bus stop. This demos how easy it is to leverage our existing APIs and resources in order to support new platforms and development.
This project visualizes health data within the Atlanta metro region. Although some research about health inequities among this region has occurred, it typically is based on county-level data. In order to have a better understanding of health inequities and disparities in our home area, a city profile for Atlanta should be established. This project has created an interactive visualization of data such as rates of teen pregnancies, low birthweight babies, etc. The system allows the viewer to explore correlations among the different variables.
Heart Sense takes biometric data from participants and produces captivating visualizations as their bodies react to visual stimuli.
Website:
Data analysis novices often encounter barriers in executing low-level operations for pairwise comparisons. They may also run into barriers in interpreting the artefacts (e.g., visualizations) created as a result of the operations. We developed Duet, a visual analysis system designed to help data analysis novices conduct pairwise comparisons by addressing execution and interpretation barriers. To reduce the barriers in executing low-level operations during pairwise comparison, Duet employs minimal specification: when one object group (i.e.
User interaction is central to the data analysis process fostered by interactive visual analytic interfaces. However, in many current systems, user interaction is represented as an ephemeral action taken by a user that moves the system from one state to another. User interactions are quantitative bits of the analytic dialog between people, the system, and the data - and when modeled - can be tactfully integrated into visual analytic systems. We propose a library to help researchers and developers capture, interpret, and model interactions in web-based visual analytic tools.
The IMAGINE (Interactive Media Architecture Group in Education) Lab is composed of a group of researchers and students with a mission of serving the Architecture/Engineering/Construction (AEC) community by: 1.
Public transportation has always been a crucial component of many metropolitan cities around the world. We’ve observed that a great amount of people around us tend to avoid taking MARTA, with one of the most common reasons being personal safety concerns. The goal of this project is to 1) Understand what factors affect riders’ perceived safety on MARTA rail system and 2) Devise a solution to increase perceived safety on MARTA rail system. Based on our research, we created a crowdsourcing prototype that helps users to report incidents and identify the population density of the incoming trains.
Data Journalists often report visualizations or statistics of data as part of the stories they tell. However, this data often comes in formats that are hard to use. This project aims to build tools to help journalists extract data from documents so that it can be easily visualized and analyzed.
At the Information Interfaces Lab, computing technologies are developed that help people take advantage of information to enrich their lives.
Visualization has an important role in science and technology. People rely on visualizations to better understand problems they have to solve. Information visualization has recently increased its domain, from being used for representations of business data, to more political and social uses via groups like visualizing.org and infosthetics.com. In parallel with this growth, we have seen the widespread acceptance of mobile technology by masses. Mobile phones, today, are being used for everything from email to ticketing and web browsing to watching videos.
In this project, we have designed visualizations that show the recent history of a team's draft selections as well as each team's regular season and playoff history. Our goal is to provide an easy-to-browse and -understand interface for exploring the data and learning about teams' pasts.
We were inspired by the COVID-19 pandemic, in which a lot of people work remotely or design their own work schedules. While flexibility is good, it is also accompanied by people being unsure of their actual productivity. So we wanted to help students and working adults who have concerns about their productivity to accurately reflect and understand their productivity.
End-user interaction with machine learning based systems will result in new usability challenges for the fields of human computer interaction and machine learning. With machine learning systems becoming more and more prevalent, it is important that systems are properly designed in a user-centered manner. In order to address these challenges, the most relied upon usability inspection method-the heuristic evaluation-must be adapted for the unique end-user experiences that interactive machine learning presents. This project introduces ten heuristics for interactive machine learning.
ITDV
ITDV
Lab:
The Interactive Topographic Data Visualizer (ITDV) is a system that facilitates a group of people to share in the exploration of geospatial solutions on an interactive projection mapped surface.
Combining querying and pattern mining for event sequence exploration
Many types of investigators routinely perform analysis that involves large collections of documents. The Jigsaw system helps investigative analysts with reasoning and sense-making in such scenarios. Jigsaw acts like a visual index onto a document collection. It first analyzes the documents, identifies entities, clusters related documents, analyzes sentiment, and summarizes each document. Next, it provides multiple visualizations of the documents, entities within, and the analysis results.
We present KnowledgeVIS, a visual analytics tool in your browser for exploring and browsing relationships that language models (LMs) have learned by identifying, comparing, and summarizing LM predictions
Website:
We conduct 3 in-lab and one crowd-sourced experiment in the domains of politics and movies. We find mixed results that interaction traces (visual scents of historical interactions) can increase awareness of bias and impact interactive behavior and decision making.
The Local Data Design Lab is focused on bridging the substantial divide between two complimentary, but largely disconnected areas of work: data studies and data visualization.
Business analysts create logomaps in order to better understand and communicate trends in the world of business. Humans can intuitively make sense of these maps, while computers struggle to extract the same knowledge. Using computer vision and human-in-the-loop machine learning, this research aims to create tools and methods for automating knowledge extraction from graphical logomaps.
Lumos is a visual data analysis tool that captures and shows the interaction history with data to increase awareness of such analytic behaviors. Using in-situ (at the place of interaction) and ex-situ (in an external view) visualization techniques, Lumos provides real-time feedback to users for them to reflect on their activities. For example, Lumos highlights datapoints that have been previously examined in the same visualization (in-situ) and also overlays them on the underlying data distribution (i.e., baseline distribution) in a separate visualization (ex-situ).
Website:
We draw upon insights from design-based research in developing the TADA (Tangible Data) system, in order to understand how data literacy, data mapping, and visualization might be aided through the translation of data into dynamic learner-created sculptures. Through the PaperMech medium, we are also able to see how storytelling and creation might be used to further understand the information in the data. We describe workshop design and discuss how the TADA system could be a useful tool in the future of 6-8th grade data science pedagogy.
We draw upon insights from design-based research in developing the TADA (Tangible Data) system, in order to understand how data literacy, data mapping, and visualization might be aided through the translation of data into dynamic learner-created sculptures. Through the PaperMech medium, we are also able to see how storytelling and creation might be used to further understand the information in the data. We describe workshop design and discuss how the TADA system could be a useful tool in the future of 6-8th grade data science pedagogy.
Examining COVID-19 Crisis Visualizations and Information Behaviors
The tool enables users to engage in structured information foraging and save interesting information scraps they encounter online. The collected information scraps are then made available in an environment that supports processing these information scraps to answer the questions users have about them.
Website:
Research, on current equine technologies, shows that limited exploration is being done in the space of wearable technology. MR ED is a research project focused on designing wearable tech for horses & riders.
Website:
The goal of this study is to learn about how the MS Human Computer Interaction program (HCI) at Georgia Tech can be tailored to suit students' specific needs and goals. Students entering the program would benefit from a greater understanding of how to move through the program to gain the skills needed for their desired roles post graduation. In order to form a complete picture of what student needs are, the research team will collect data about the current MS HCI program structure, current industry expectations of HCI grads, and current student expectations.
Data visualization tool to identify stressful factors for cyclist to help city planners make better decisions on cycling infrastructure in Atlanta
This demo shows a system called Dust and Magnet (DnM) that is a general purpose data visualization system. DnM represents data items as iron dust. Each attribute of the data then is a magnet. The system is implemented on a large multi-touch display where the analyst can deploy magnets and drag them around the view. Data points will then be attracted more strongly or weakly depending on that data item's value of the attribute represented by each magnet. This system provides a very hands-on, visceral data exploration experience.
The motion of our articulators is responsible for generating speech. To re-learn how to speak intelligibly after a brain injury, practicing enunciation when learning a new language, or change our native accent to fit in a new place or required as part of our job, accessing and correcting our articulators' motion is crucial. Few practical solutions exist in the market, and our system is filling this gap by providing an affordable, unobtrusive, portable and user-friendly solution to visualize and generate feedback to the user about their speech performance (e.g., tongue and lips gesture).
We designed and developed two interactive visualization systems, NBA GameViz and NBA LineupViz. NBA GameViz assists sportswriters in post-game coverage, and NBA Lineup Viz provides sophisticated evaluation regarding lineups performances.
NL4DV is a Python package that takes as input a tabular dataset and a natural language query about that dataset. In response, the toolkit returns an analytic specification comprising data attributes, analytic tasks, and a list of visualizations (Vega-Lite specifications) relevant to the input query.
Website:
We investigate how users perceive socioeconomic disparities in the United States before and after visualizing and interacting with a standalone county-level US choropleth map and a choropleth map of the US that is `linked' to a map of the world.
We investigate how users perceive socioeconomic disparities in the United States before and after visualizing and interacting with a standalone county-level US choropleth map and a choropleth map of the US that is `linked' to a map of the world. We measure socioeconomic properties using indicators related to education attainment, income inequality, life expectancy, incarceration, obesity and prevalence of HIV. We test whether perceive inequality across the US differently if they can contextualize variation alongside the same indicators at the global scale using an A/B user study.
Visualizing sets to reveal relationships between constituent elements is a complex representational problem. Recent research presents several automated placement and grouping techniques to highlight connections between set elements. However, these techniques do not scale well for sets with cardinality greater than one hundred elements. We present OnSet, an interactive, scalable visualization technique for representing large-scale binary set data.
Data visualization systems have predominantly been developed for WIMP-based direct manipulation interfaces. Only recently have other forms of interaction begun to appear, such as natural language or touch-based interaction, though usually operating only independently. Prior evaluations of natural language interfaces for visualization have indicated potential value in combining direct manipulation and natural language as complementary interaction techniques. Unfortunately, however, little work has been done in exploring such multimodal visualization interfaces.
Informing the planning of hikes and trail walks through immersive experiences right from your screen at home.
Designing a participative approach for multiple users to control their smart home.
Theories of Quantum Mechanics(QM) have been central to the philosophical and technological advances in physics and related fields. Some of the most important aspects of these theories are outside the bounds of human experience, predominantly explained and taught drawing on abstract mathematical formulas.
People often rank and order data items as a vital part of making decisions. Multi-attribute ranking systems are a common tool used to make such data-driven decisions. These systems are often table-based tools that can produce rankings based on numerical weights that a user assigns to each attribute, where the weight represents how important the user believes an attribute is to their decision. These systems assume that users are able to quantify their conceptual understanding of how important particular attributes are; however, this is not always the case.
Healthcare delivery processes are complex activities that span organizational, spatial, and temporal boundaries. Systemic insights, redesign, and improvements are consequently difficult to achieve. Using existing digital healthcare data, we are developing a data-driven methodology, fusing computational systems modeling, data mining, and interactive visualization, to identify, describe, and visualize healthcare delivery processes. Our system will help providers (e.g. physicians, nurses, etc.) and strategic decision makers (e.g.
With the remarkable increase of data, novel tools and metrics are needed for comprehensive and systemic analyses of converging business ecosystems. Knowledge discovery is the computational process of identifying valid, novel, interesting, potentially useful and ultimately understandable patterns in data. The objective of our study is twofold. First, we introduce the emerging domain of "big" business ecosystem data. Second, we describe the success and challenges that we encountered in analyzing this data using state of the art analytics and visualization techniques.
Psi and Delta is a collaborative classroom videogame that enables students to experience the world of the very small, together.
Website:
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.
Website:
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
Diagnostic radiology reports are increasingly being made available to patients and their family members. However, these reports are not typically comprehensible to lay, recipients, impeding effective communication about report findings. Rapport is a prototype system that aims to facilitate communication about radiology imaging findings among pediatric patients, their family members and clinicians in the clinical setting.
Website:
The streetcars run in the heart of downtown. They are subject to unstable operating conditions caused by traffic congestion, basketball games, and obstructed right-of-way. These perturbations make the Atlanta Streetcar prone to streetcar-bunching, which causes undue passenger wait and crowding. We have developped a real-time dispatching method that considers every streetcar on the route to dispatch them with even headways, while maintaining a high frequency of service.
In order to prevent spinal surgery, scoliosis patients often are prescribed to wear a corrective brace to keep the spine from continuing to develop curvature. While largely successful if followed, many patients do not comply with wearing their brace for the correct amount of time. Our system attempts to improve compliance with patients by providing real-time data of their wear time in a meaningful way.
Redesigning the Receipts Through An Balanced Study of HCI and Design
Remote research often provides flawed insights into the lifestyles and workplaces of the interviewees. How might research tools evolve to get better contextual insights from users without losing the inherent advantages of remote research?
Large, multisensor datasets are available covering a large portion of Mars. Analysis and display of these datasets are currently in use for path planning tools that provide a precise, low-level visualization that fosters precision planning for Rover Planners at NASA Jet Propulsion Laboratory. However, these visualizations do not foster path planning at a higher level of abstraction. In addition, planning a path uses a non-intuitive process of generating rover commands, simulating them, visualizing the results, and then tweaking the commands until the path looks correct.
Spatio-temporal data is often displayed using regional aggregation or heatmaps, which are useful for exploring large distributed trends or working to unearth the cause of more localized behavior. For individual users that live and work in the region, however, these representations are inaccessible and difficult to put into practice. We present a new technique for exploring spatio-temporal data as personal routes through a geographic area.
Recently featured in Georgia County Government magazine, the SayWhyPoll mobile app enables elected officials, civic leaders, and media producers to engage with constituents and audiences either remotely or face-to-face using surveys that tightly couple close-ended survey items with rich media, such as video. The SayWhyPoll is designed to increase opportunities for public debate on civic issues, but is also suitable for pure entertainment topics, such as sports and lifestyle.
SCADE is a visual text analytic tool. The goal of the project is to help analysts make sense of a larger number of text document while tracking the analyst's provenance.
The growing popularity of social media makes it increasingly difficult to keep up with the huge volumes of information they produce. We present SentenTree, a novel visualization technique that helps people gain a quick understanding of the key concepts and opinions expressed in a given social media text set. SentenTree can be used by both casual social media users and professional analysts.
Can sensor-instrumented toys be used to monitor health and improve enrichment for sea otters at the Georgia Aquarium? This project looks at the design requirements of computer-driven otter enrichment devices and how we can help otter trainers derive meaningful health insights from the data pulled from such devices.
Visualization tools for spatiotemporal data utilize map-based representations to help a user understand trends and outliers within a given region over time. Multitouch visualization tools allow us to recreate many of the capabilities of sketching directly on maps while still taking advantage of computational models of public safety. We will be demonstrating SpaceSketch, a multitouch approach to spatiotemporal visualization. Visitors will be allowed to explore crime and transmit data in the city of Atlanta using our high-resolution Surface Hub Interface.
Sparse Tangibles investigates the use of novel tangible and gestural interactions for making sense of large biological datasets. Our current prototype employs active tangibles in combination with a large multi-touch tabletop displays to navigate and visualize gene regulatory network data from the BioGrid database.
In this project, I designed and developed a research dashboard for researchers to better understand Speech on Twitter.
Dimensionality reduction (DR) is often used for exploring the evolution patterns of a collection of dynamic ego-networks. However, DR often lacks flexibility: as analysts' questions evolve during data exploration, the low-dimensional projection remains static, limiting the depth of exploration. To address the inflexibility of DR, we designed a data transformation pipeline which enables analysts to transform dynamic ego-networks into event sequences for steering MDS to create different scatterplots of dynamic ego-networks.
Can locative media (Augmented and Mixed Reality, web applications, and social networking) serve as a platform for preservation of cultural heritage, informal education, and civic engagement? This is the question at the heart of the Auburn Avenue Research Project, a project that brings together researchers from a variety of disciplines – including media theory, design studies, and human-computer interaction – to engage the above question in theory and practice.
TSynlab explores emerging modalities in new media. Our research focuses on tangible interaction and sensing technologies that support creative expression bridging the physical and digital worlds. Applications range across media arts, entertainment and educational domains.
This project is focused on designing a tangible and auditory periodic table educational tool for visually impaired and sighted students.
Reimagining a lost nineteenth-century haptic visualization scheme
In collections of scientific and cultural history that are too big to see, metadata act as virtual handles for rare and delicate artifacts from the past. At the Arnold Arboretum, a collection of long-lived trees, vines, and shrubs managed by Harvard University, landscapes from around the world and across time are stitched together by metadata. However, metadata are worthy of study themselves. Created in varied social and technological eras, they register the organizational structures and values of their time.
The Light Orchard is an interactive installation that invites people to walk into a grove of futuristic trees, lit with color. The trees are aware of the presence of people in their space, and can respond in many different ways. User can play different games, watch animations, and work together with different simulations, that allow them to easily collaborate, learn, and play together.
Website:
Using machine learning to help scholars find new research sources
Website:
Investing in the financial market has never been easier with online brokers and market news only a tap away for individual investors. However, behavioral finance theories show that individual investors often suffer from decision biases that could negatively impact their investment return. In this project, I will leverage data visualization on personal trade history data to help investors improve their decision making.
Abstracting of the city's gaze using critical design
We are interested in ubiquitous computing and the research issues involved in building and evaluating ubicomp applications and services that impact our lives. Much of our work is situated in settings of everyday activity, such as the classroom, the office and the home.
This project focuses on building technology for law enforcement working on human trafficking cases. We leverage available data to build tools that help law enforcement identify potential victims and collaborate with partners to best intervene in these cases.
Website:
The proposed research will create Visual Analytics by Demonstration prototypes, generalizable toolkits, and demonstration primitives to foster exploration and discovery in visual analytics.
Our goal is to help people make sense of data. We research and develop interactive visualizations that couple machine learning with visual interfaces of data for exploration and sensemaking.
The Visual Policy Initiative aims to transform complex policy issues into easy to understand data visualizations using empirically-derived evidence. The Visual Policy team is comprised of a group of researchers from both public policy and digital media. Through this collaborative effort, we aim to transform complex policy issues into easy to understand data visualizations using empirically-derived evidence.
Visualization Journalism is focused on developing an interface and graphical metalanguage for massive multimodal news datasets. Such datasets are increasingly available, but for copyright reasons, they cannot be made entirely open to the public. The project seeks to offer an abstracted and legal representation of news data, to enable comparative, cooperative and computer-supported analysis of trends across news events and networks.
This study describes a data-driven visualization approach to the systemic study of innovations in global supply chain networks. We demonstrate its applicability with illustrative examples of real-world supply chain in the electronics industry. Our visualization approach enhances the hypothesis-generating process as it can reveal important clusters, patterns, trends, and outliers in the networks.
The project's goal is to help people to identify some of the variables involved in air traffic safety, and understand that air traffic safety relies on both technology and the people who control it. Hopefully, it will be illuminating to anyone concerned with air traffic safety.
The design and production of complex engineered systems (CES) require analysis of massive amounts of detailed information, including data on products and materials, engineering designs, manufacturing specifications, supply chain and delivery data, and changing customer needs. Visual analytics promises to offer tools and methods that will help stakeholders interactively explore, discover, and make sense of the underlying data.
A visualization system for portraying the jumps in men's and ladies' single figure skating programs. Data is from the International Skating Union's score tables for world championships in the last 6 years. The objective is to better interpret the score tables by visualizing the program composition of top skaters, as well as showing a trend of the sport in general.
The PGA Tour provides an extensive data collection of information about players' performance and individual shots over the past few years. This collection is called ShotLink data. In this project we are building an interactive visualization system that will allow the viewer to easily browse and explore the golf shot data to learn more about the performance of all the players on tour. The system presents a variety of different statistics including scoring, driving accuracy, greens in regulation, putting, and so on.
We have created a visual interface to explore the history of the top 100 U.S. golf course rankings from Golf Digest and Golf Magazines. A viewer can explore the courses geographically via a map or through the individual ordered lists from the magazines. The system shows how each course's ranking has changed over the years, and it allows the viewer to explore courses by particular architects.
Improve the usabilty of a vaccine data visualization, including ways to visualize uncertainty
We developed a visual analytics system, VitaLITy, to promote serendipitous discovery of academic papers wherein users may “stumble upon” relevant literature, when other search approaches may fail. VitaLITy (1) utilizes transformer language models to help users find semantically similar papers given a list of seed paper(s) or a working abstract, (2) visualizes the embedding space in an interactive 2-D scatterplot, and (3) summarizes meta information about the paper corpus (e.g., keywords, co-authors, citation counts, and publication year).
Website:
VROOM
VROOM
Lab:
This project investigates the effect of providing this information on users’ virtual experience, wellbeing, and behavior in the dynamic real-world environment.
People increasingly rely on visual representations of information to explore and make sense of data. However, people have inherent biases that often lead to errors and inefficiencies in the decision making process. The goal of our research is to help people make better decisions while exploring and analyzing data using visualizations. We introduce computational methods for quantifying an analyst’s biases based on their interactions in the visualization. Using that information, we illustrate ways to modify or design new visualization systems that mitigate biased decision making.
Data visualization can be an important guiding force in scientific debates and casual discussions alike. Bringing data visibly into the world can inform and bring attention to critical issues, as well as help us develop a more personal relationship with the data. This project aims to promote awareness and stimulate discussion about climate change through visualizing personal carbon footprint data on clothing. It explores the placement of visualizations in the social sphere, as well as revealing unseen individual and systematic responsibility for carbon emissions.
The Wearable Technology Designer's Web Tool is a tool that leads designers / developers through a series of questions about their wearable technology project. The questions are designed to force designers to think of all options at the beginning of a project, illuminating opportunities and shortcomings in accessibility caused by decisions along the way.
