The Computational Thematic Analysis Toolkit

As online communities have grown, Computational Social Science has rapidly developed new techniques to study them. However, these techniques require researchers to become experts in a wide variety of tools in addition to qualitative and computational research methods. Studying online communities also requires researchers to constantly navigate highly contextual ethical and transparency considerations when engaging with data, such as respecting their members' privacy when discussing sensitive or stigmatized topics. To overcome these challenges, we developed the Computational Thematic Analysis Toolkit, a modular software package that supports analysis of online communities by combining aspects of reflexive thematic analysis with computational techniques. Our toolkit demonstrates how common analysis tasks like data collection, cleaning and filtering, modelling and sampling, and coding can be implemented within a single visual interface, and how that interface can encourage researchers to manage ethical and transparency considerations throughout their research process.

Interfaces for Searching and Triaging Large Document Sets: An Ontology-Supported Visual Analytics Approach

We investigate the design of ontology-supported, progressively disclosed visual analytics interfaces for searching and triaging large document sets. The goal is to distill a set of criteria that can help guide the design of such systems. We begin with a background of information search, triage, machine learning, and ontologies. We review research on the multi-stage information-seeking process to distill the criteria. To demonstrate their utility, we apply the criteria to the design of a prototype visual analytics interface: VisualQUEST (Visual interface for QUEry, Search, and Triage). VisualQUEST allows users to plug-and-play document sets and expert-defined ontology files within a domain-independent environment for multi-stage information search and triage tasks. We describe VisualQUEST through a functional workflow and culminate with a discussion of ongoing formative evaluations, limitations, future work, and summary.

Mapping Chicago Schools

The idea of the “Chicago school” is first attested as early as 1850, in medicine. It symbolized a way of thinking that united theory and practice. This way of thinking was later adapted to other fields of study, including architecture, design, social science, economics, theology, literature, and beyond. As a result of such adaptation, various “Chicago schools” emerged in these different fields of study. Most of these Chicago schools were seen as attempts to unite theory and practice. The HathiTrust holds a broad range of books and periodicals that mention one or another “Chicago school.” Here, I create maps for a selection of Chicago schools. The maps show in which contexts the Chicago schools have been mentioned. All information is considered that is published on the same page as the mention of the “Chicago school”. The visual interface shown on the next slides can be found online. The graphical interface and the method of geographical information retrieval that I employ are my and my collaborators’ work, initially started as part of my ACS project with the HathiTrust Research Center (2016).

Collaborative Software Visualization in Co-located Environments

<p>Most software visualization systems and tools are designed from a single-user perspective and are bound to the desktop and IDEs. These design decisions do not allow users to analyse software collaboratively or to easily interact and navigate visualizations within a co-located environment at the same time. This thesis presents an exploratory study of collaborative software visualization using multi-touch tables in a co-located environment. The thesis contributes a richer understanding of how pairs of developers make use of shared visualizations on large multi-touch tables to gain insight into the design of software systems. We designed a collaborative software visualization application, called Source-Vis, that contained a suite of 13 visualization techniques adapted for multi-touch interaction. We built two large multi-touch tables (28 and 48 inches) following existing hardware designs, to explore and evaluate SourceVis. We then conducted both qualitative and quantitative user studies, culminating in a study of 44 professional software developers working in pairs. We found that pairs preferred joint group work, used a variety of coupling styles, and made many transitions between coupling and arrangement styles. For collaborative group work we recommend designing for joint group work over parallel individual work, supporting a flexible variety of coupling styles, and supporting fluid transitions between coupling and arrangement styles. We found that the preferred style for joint group work was closely coupled and arranged side by side. We found some global functionally was not easily accessible. We found some of the user interactions and visual interface elements were not designed consistently. For the design of collaborative software visualizations we recommend designing visualizations for closely coupled arrangements with rotation features, providing functionality in the appropriate locality, and providing consistent user interactions and visual interface design. We found sometimes visualization windows overlapped each other and text was hard to read in windows. We found when pairs were performing joint group work the size of the table was appropriate but not for parallel individual. We found that because the table could not differentiate between different simultaneous users that some pair interactions were limited. For the design of multi-touch tables we recommend providing a high resolution workspace, providing appropriate table space, and differentiating between simultaneous user interactions.</p>

Collaborative Software Visualization in Co-located Environments

<p>Most software visualization systems and tools are designed from a single-user perspective and are bound to the desktop and IDEs. These design decisions do not allow users to analyse software collaboratively or to easily interact and navigate visualizations within a co-located environment at the same time. This thesis presents an exploratory study of collaborative software visualization using multi-touch tables in a co-located environment. The thesis contributes a richer understanding of how pairs of developers make use of shared visualizations on large multi-touch tables to gain insight into the design of software systems. We designed a collaborative software visualization application, called Source-Vis, that contained a suite of 13 visualization techniques adapted for multi-touch interaction. We built two large multi-touch tables (28 and 48 inches) following existing hardware designs, to explore and evaluate SourceVis. We then conducted both qualitative and quantitative user studies, culminating in a study of 44 professional software developers working in pairs. We found that pairs preferred joint group work, used a variety of coupling styles, and made many transitions between coupling and arrangement styles. For collaborative group work we recommend designing for joint group work over parallel individual work, supporting a flexible variety of coupling styles, and supporting fluid transitions between coupling and arrangement styles. We found that the preferred style for joint group work was closely coupled and arranged side by side. We found some global functionally was not easily accessible. We found some of the user interactions and visual interface elements were not designed consistently. For the design of collaborative software visualizations we recommend designing visualizations for closely coupled arrangements with rotation features, providing functionality in the appropriate locality, and providing consistent user interactions and visual interface design. We found sometimes visualization windows overlapped each other and text was hard to read in windows. We found when pairs were performing joint group work the size of the table was appropriate but not for parallel individual. We found that because the table could not differentiate between different simultaneous users that some pair interactions were limited. For the design of multi-touch tables we recommend providing a high resolution workspace, providing appropriate table space, and differentiating between simultaneous user interactions.</p>