Despite their increasing capabilities, text-to-image generative AI systems are known to produce biased, offensive, and otherwise problematic outputs. While recent advancements have supported testing and auditing of generative AI, existing auditing methods still face challenges in supporting effectively explore the vast space of AI-generated outputs in a structured way. To address this gap, we conducted formative studies with five AI auditors and synthesized five design goals for supporting systematic AI audits. Based on these insights, we developed Vipera, an interactive auditing interface that employs multiple visual cues including a scene graph to facilitate image sensemaking and inspire auditors to explore and hierarchically organize the auditing criteria. Additionally, Vipera leverages LLM-powered suggestions to facilitate exploration of unexplored auditing directions. Through a controlled experiment with 24 participants experienced in AI auditing, we demonstrate Vipera's effectiveness in helping auditors navigate large AI output spaces and organize their analyses while engaging with diverse criteria.

Artificial intelligence (AI)-based decision support systems can be highly accurate yet still fail to support users or improve decisions. Existing theories of AI-assisted decision-making focus on calibrating reliance on AI advice, leaving it unclear how different system designs might influence the reasoning processes underneath. We address this gap by reconsidering AI interfaces as collections of intelligent reasoning cues: discrete pieces of AI information that can individually influence decision-making. We then explore the roles of eight types of reasoning cues in a high-stakes clinical decision (treating patients with sepsis in intensive care). Through contextual inquiries with six teams and a think-aloud study with 25 physicians, we find that reasoning cues have distinct patterns of influence that can directly inform design. Our results also suggest that reasoning cues should prioritize tasks with high variability and discretion, adapt to ensure compatibility with evolving decision needs, and provide complementary, rigorous insights on complex cases.

Data profiling plays a critical role in understanding the structure of complex datasets and supporting numerous downstream tasks, such as social media analytics and financial fraud detection. While existing research predominantly focuses on structured data formats, a substantial portion of semi-structured textual data still requires ad-hoc and arduous manual profiling to extract and comprehend its internal structures. In this work, we propose StructVizor, an interactive profiling system that facilitates sensemaking and transformation of semi-structured textual data. Our tool mainly addresses two challenges: a) extracting and visualizing the diverse structural patterns within data, such as how information is organized or related, and b) enabling users to efficiently perform various wrangling operations on textual data. Through automatic data parsing and structure mining, StructVizor enables visual analytics of structural patterns, while incorporating novel interactions to enable profile-based data wrangling. A comparative user study involving 12 participants demonstrates the system's usability and its effectiveness in supporting exploratory data analysis and transformation tasks.

Analyzing data subgroups is a common data science task to build intuition about a dataset and identify areas to improve model performance. However, subgroup analysis is prohibitively difficult in datasets with many features, and existing tools limit unexpected discoveries by relying on user-defined or static subgroups. We propose exploratory subgroup analysis as a set of tasks in which practitioners discover, evaluate, and curate interesting subgroups to build understanding about datasets and models. To support these tasks we introduce Divisi, an interactive notebook-based tool underpinned by a fast approximate subgroup discovery algorithm. Divisi's interface allows data scientists to interactively re-rank and refine subgroups and to visualize their overlap and coverage in the novel Subgroup Map. Through a think-aloud study with 13 practitioners, we find that Divisi can help uncover surprising patterns in data features and their interactions, and that it encourages more thorough exploration of subtypes in complex data.

Temporal predictive models have the potential to improve decisions in health care, public services, and other domains, yet they often fail to effectively support decision-makers. Prior literature shows that many misalignments between model behavior and decision-makers' expectations stem from issues of model specification, namely how, when, and for whom predictions are made. However, model specifications for predictive tasks are highly technical and difficult for non-data-scientist stakeholders to interpret and critique. To address this challenge we developed Tempo, an interactive system that helps data scientists and domain experts collaboratively iterate on model specifications. Using Tempo's simple yet precise temporal query language, data scientists can quickly prototype specifications with greater transparency about pre-processing choices. Moreover, domain experts can assess performance within data subgroups to validate that models behave as expected. Through three case studies, we demonstrate how Tempo helps multidisciplinary teams quickly prune infeasible specifications and identify more promising directions to explore.

Drug overdose deaths, including those due to prescription opioids, represent a critical public health issue in the United States and worldwide. Artificial intelligence (AI) approaches have been developed and deployed to help prescribers assess a patient's risk for overdose-related death, but it is unknown whether public health experts can leverage similar predictions to make local resource allocation decisions more effectively. In this work, we evaluated how AI-based overdose risk assessment could be used to inform local public health decisions using a working prototype system. Experts from three health departments, of varying locations and sizes with respect to staff and population served, were receptive to the potential benefits of algorithmic risk prediction and of using AI-augmented visualization to connect across data sources. However, they also expressed concerns about whether the risk prediction model's formulation and underlying data would match the state of the overdose epidemic as it evolved in their specific locations. Our findings extend those of other studies on algorithmic systems in the public sector, and they present opportunities for future human-AI collaborative tools to support decision-making in local, time-varying contexts.

With the increasing use of AI, recent research in human–computer interaction explores Explainable AI (XAI) to make AI advice more interpretable. While research addresses the effects of incorrect AI advice on AI-assisted decision-making, the impact of incorrect explanations is neglected so far. Additionally, recent work shows that not only different explanation modalities impact decision-makers, but also human factors play a critical role. To analyze relevant phenomena influencing AI-assisted decision-making, this work explores the impacting factors by conceptualizing theories of appropriate reliance and taking the first steps toward empirical evidence. We show that humans’ reliance on AI and the human–AI team performance are impacted by imperfect XAI in a study with 136 participants. Additionally, we find that cognitive styles affect decision-making in different explanation modalities. Hence, we shed light on diverse factors that impact human–AI collaboration and provide guidelines for designers to tailor such human–AI collaboration systems to individuals’ needs.

This January 2025 issue of the IEEE Transactions on Visualization and Computer Graphics (TVCG) contains the proceedings of IEEE VIS 2024, held on October 1318 October, 2024 in St. Pete Beach, Florida, USA, with the three General Chairs Paul Rosen (University of Utah), Kristi Potter (U.S. National Renewable Energy Laboratory), and Remco Chang (Tufts University). With IEEE VIS 2024, the conference series is in its 35th year.

Advances in artificial intelligence (AI) have enabled unprecedented capabilities, yet innovation teams struggle when envisioning AI concepts. Data science teams think of innovations users do not want, while domain experts think of innovations that cannot be built. A lack of effective ideation seems to be a breakdown point. How might multidisciplinary teams identify buildable and desirable use cases? This paper presents a first hand account of ideating AI concepts to improve critical care medicine. As a team of data scientists, clinicians, and HCI researchers, we conducted a series of design workshops to explore more effective approaches to AI concept ideation and problem formulation. We detail our process, the challenges we encountered, and practices and artifacts that proved effective. We discuss the research implications for improved collaboration and stakeholder engagement, and discuss the role HCI might play in reducing the high failure rate experienced in AI innovation.

Many papers make claims about specific visualization techniques that are said to enhance or calibrate trust in AI systems. But a design choice that enhances trust in some cases appears to damage it in others. In this paper, we explore this inherent duality through an analogy with "knobs". Turning a knob too far in one direction may result in under-trust, too far in the other, over-trust or, turned up further still, in a confusing distortion. While the designs or so-called "knobs" are not inherently evil, they can be misused or used in an adversarial context and thereby manipulated to mislead users or promote unwarranted levels of trust in AI systems. When a visualization that has no meaningful connection with the underlying model or data is employed to enhance trust, we refer to the result as "trust junk." From a review of 65 papers, we identify nine commonly made claims about trust calibration. We synthesize them into a framework of knobs that can be used for good or "evil," and distill our findings into observed pitfalls for the responsible design of human-AI systems.

Custom animated visualizations of large, complex datasets are helpful across many domains, but they are hard to develop. Much of the difficulty arises from maintaining visualization state across many animated graphical elements that may change in number over time. We contribute Counterpoint, a framework for state management designed to help implement such visualizations in JavaScript. Using Counterpoint, developers can manipulate large collections of marks with reactive attributes that are easy to render in scalable APIs such as Canvas and WebGL. Counterpoint also helps orchestrate the entry and exit of graphical elements using the concept of a rendering "stage." Through a performance evaluation, we show that Counterpoint adds minimal overhead over current high-performance rendering techniques while simplifying implementation. We provide two examples of visualizations created using Counterpoint that illustrate its flexibility and compatibility with other visualization toolkits as well as considerations for users with disabilities. Counterpoint is open-source and available at https://github.com/cmudig/counterpoint.

Statistical practices such as building regression models or running hypothesis tests rely on following rigorous procedures of steps and verifying assumptions on data to produce valid results. However, common statistical tools do not verify users’ decision choices and provide low-level statistical functions without instructions on the whole analysis practice. Users can easily misuse analysis methods, potentially decreasing the validity of results. To address this problem, we introduce GuidedStats, an interactive interface within computational notebooks that encapsulates guidance, models, visualization, and exportable results into interactive workflows. It breaks down typical analysis processes, such as linear regression and two-sample T-tests, into interactive steps supplemented with automatic visualizations and explanations for step-wise evaluation. Users can iterate on input choices to refine their models, while recommended actions and exports allow the user to continue their analysis in code. Case studies show how GuidedStats offers valuable instructions for conducting fluid statistical analyses while finding possible assumption violations in the underlying data, supporting flexible and accurate statistical analyses.

Explainability techniques are rapidly being developed to improve human-AI decision-making across various cooperative work settings. Consequently, previous research has evaluated how decision-makers collaborate with imperfect AI by investigating appropriate reliance and task performance with the aim of designing more human-centered computer-supported collaborative tools. Several human-centered explainable AI (XAI) techniques have been proposed in hopes of improving decision-makers' collaboration with AI; however, these techniques are grounded in findings from previous studies that primarily focus on the impact of incorrect AI advice. Few studies acknowledge the possibility of the explanations being incorrect even if the AI advice is correct. Thus, it is crucial to understand how imperfect XAI affects human-AI decision-making. In this work, we contribute a robust, mixed-methods user study with 136 participants to evaluate how incorrect explanations influence humans' decision-making behavior in a bird species identification task, taking into account their level of expertise and an explanation's level of assertiveness. Our findings reveal the influence of imperfect XAI and humans' level of expertise on their reliance on AI and human-AI team performance. We also discuss how explanations can deceive decision-makers during human-AI collaboration. Hence, we shed light on the impacts of imperfect XAI in the field of computer-supported cooperative work and provide guidelines for designers of human-AI collaboration systems.

This paper introduces an innovative methodology for producing high-quality 3D lung CT images guided by textual information. While diffusion-based generative models are increasingly used in medical imaging, current state-of-the-art approaches are limited to low-resolution outputs and underutilize radiology reports' abundant information. The radiology reports can enhance the generation process by providing additional guidance and offering fine-grained control over the synthesis of images. Nevertheless, expanding text-guided generation to high-resolution 3D images poses significant memory and anatomical detail-preserving challenges. Addressing the memory issue, we introduce a hierarchical scheme that uses a modified UNet architecture. We start by synthesizing low-resolution images conditioned on the text, serving as a foundation for subsequent generators for complete volumetric data. To ensure the anatomical plausibility of the generated samples, we provide further guidance by generating vascular, airway, and lobular segmentation masks in conjunction with the CT images. The model demonstrates the capability to use textual input and segmentation tasks to generate synthesized images. Algorithmic comparative assessments and blind evaluations conducted by 10 board-certified radiologists indicate that our approach exhibits superior performance compared to the most advanced models based on GAN and diffusion techniques, especially in accurately retaining crucial anatomical features such as fissure lines and airways. This innovation introduces novel possibilities. This study focuses on two main objectives: (1) the development of a method for creating images based on textual prompts and anatomical components, and (2) the capability to generate new images conditioning on anatomical elements. The advancements in image generation can be applied to enhance numerous downstream tasks.

Profiling data by plotting distributions and analyzing summary statistics is a critical step throughout data analysis. Currently, this process is manual and tedious since analysts must write extra code to examine their data after every transformation. This inefficiency may lead to data scientists profiling their data infrequently, rather than after each transformation, making it easy for them to miss important errors or insights. We propose continuous data profiling as a process that allows analysts to immediately see interactive visual summaries of their data throughout their data analysis to facilitate fast and thorough analysis. Our system, AutoProfiler, presents three ways to support continuous data profiling: (1) it automatically displays data distributions and summary statistics to facilitate data comprehension; (2) it is live, so visualizations are always accessible and update automatically as the data updates; (3) it supports follow up analysis and documentation by authoring code for the user in the notebook. In a user study with 16 participants, we evaluate two versions of our system that integrate different levels of automation: both automatically show data profiles and facilitate code authoring, however, one version updates reactively ("live") and the other updates only on demand ("dead"). We find that both tools, dead or alive, facilitate insight discovery with 91% of user-generated insights originating from the tools rather than manual profiling code written by users. Participants found live updates intuitive and felt it helped them verify their transformations while those with on-demand profiles liked the ability to look at past visualizations. We also present a longitudinal case study on how AutoProfiler helped domain scientists find serendipitous insights about their data through automatic, live data profiles. Our results have implications for the design of future tools that offer automated data analysis support.

Machine learning models with high accuracy on test data can still produce systematic failures, such as harmful biases and safety issues, when deployed in the real world. To detect and mitigate such failures, practitioners run behavioral evaluation of their models, checking model outputs for specific types of inputs. Behavioral evaluation is important but challenging, requiring that practitioners discover real-world patterns and validate systematic failures. We conducted 18 semi-structured interviews with ML practitioners to better understand the challenges of behavioral evaluation and found that it is a collaborative, use-case-first process that is not adequately supported by existing task- and domain-specific tools. Using these findings, we designed zeno, a general-purpose framework for visualizing and testing AI systems across diverse use cases. In four case studies with participants using zeno on real-world models, we found that practitioners were able to reproduce previous manual analyses and discover new systematic failures.

Artificial intelligence (AI) in healthcare has the potential to improve patient outcomes, but clinician acceptance remains a critical barrier. We developed a novel decision support interface that provides interpretable treatment recommendations for sepsis, a life-threatening condition in which decisional uncertainty is common, treatment practices vary widely, and poor outcomes can occur even with optimal decisions. This system formed the basis of a mixed-methods study in which 24 intensive care clinicians made AI-assisted decisions on real patient cases. We found that explanations generally increased confidence in the AI, but concordance with specific recommendations varied beyond the binary acceptance or rejection described in prior work. Although clinicians sometimes ignored or trusted the AI, they also often prioritized aspects of the recommendations to follow, reject, or delay in a process we term “negotiation.” These results reveal novel barriers to adoption of treatment-focused AI tools and suggest ways to better support differing clinician perspectives.

People work with AI systems to improve their decision making, but often under- or over-rely on AI predictions and perform worse than they would have unassisted. To help people appropriately rely on AI aids, we propose showing them behavior descriptions, details of how AI systems perform on subgroups of instances. We tested the efficacy of behavior descriptions through user studies with 225 participants in three distinct domains: fake review detection, satellite image classification, and bird classification. We found that behavior descriptions can increase human-AI accuracy through two mechanisms: helping people identify AI failures and increasing people's reliance on the AI when it is more accurate. These findings highlight the importance of people's mental models in human-AI collaboration and show that informing people of high-level AI behaviors can significantly improve AI-assisted decision making.

Recent developments in explainable AI (XAI) aim to improve the transparency of black-box models. However, empirically evaluating the interpretability of these XAI techniques is still an open challenge. The most common evaluation method is algorithmic performance, but such an approach may not accurately represent how interpretable these techniques are to people. A less common but growing evaluation strategy is to leverage crowd-workers to provide feedback on multiple XAI techniques to compare them. However, these tasks often feel like work and may limit participation. We propose a novel, playful, human-centered method for evaluating XAI techniques: a Game With a Purpose (GWAP), Eye into AI, that allows researchers to collect human evaluations of XAI at scale. We provide an empirical study demonstrating how our GWAP supports evaluating and comparing the agreement between three popular XAI techniques (LIME, Grad-CAM, and Feature Visualization) and humans, as well as evaluating and comparing the interpretability of those three XAI techniques applied to a deep learning model for image classification. The data collected from Eye into AI offers convincing evidence that GWAPs can be used to evaluate and compare XAI techniques. Eye into AI is available to the public: https://dig.cmu.edu/eyeintoai/.

Artificial intelligence (AI) is increasingly being deployed in high-stakes domains, such as disaster relief and radiology, to aid practitioners during the decision-making process. Explainable AI techniques have been developed and deployed to provide users insights into why the AI made certain predictions. However, recent research suggests that these techniques may confuse or mislead users. We conducted a series of two studies to uncover strategies that humans use to explain decisions and then understand how those explanation strategies impact visual decision-making. In our first study, we elicit explanations from humans when assessing and localizing damaged buildings after natural disasters from satellite imagery and identify four core explanation strategies that humans employed. We then follow up by studying the impact of these explanation strategies by framing the explanations from Study 1 as if they were generated by AI and showing them to a different set of decision-makers performing the same task. We provide initial insights on how causal explanation strategies improve humans' accuracy and calibrate humans' reliance on AI when the AI is incorrect. However, we also find that causal explanation strategies may lead to incorrect rationalizations when AI presents a correct assessment with incorrect localization. We explore the implications of our findings for the design of human-centered explainable AI and address directions for future work.

Data scientists require rich mental models of how AI systems behave to effectively train, debug, and work with them. Despite the prevalence of AI analysis tools, there is no general theory describing how people make sense of what their models have learned. We frame this process as a form of sensemaking and derive a framework describing how data scientists develop mental models of AI behavior. To evaluate the framework, we show how existing AI analysis tools fit into this sensemaking process and use it to design AIFinnity , a system for analyzing image-and-text models. Lastly, we explored how data scientists use a tool developed with the framework through a think-aloud study with 10 data scientists tasked with using AIFinnity to pick an image captioning model. We found that AIFinnity ’s sensemaking workflow reflected participants’ mental processes and enabled them to discover and validate diverse AI behaviors.

Data-driven AI systems are increasingly used to augment human decision-making in complex, social contexts, such as social work or legal practice. Yet, most existing design knowledge regarding how to best support AI-augmented decision-making comes from studies in comparatively well-defined settings. In this paper, we present findings from design interviews with 12 social workers who use an algorithmic decision support tool (ADS) to assist their day-to-day child maltreatment screening decisions. We generated a range of design concepts, each envisioning different ways of redesigning or augmenting the ADS interface. Overall, workers desired ways to understand the risk score and incorporate contextual knowledge, which move beyond existing notions of AI interpretability. Conversations around our design concepts also surfaced more fundamental concerns around the assumptions underlying statistical prediction, such as inference based on similar historical cases and statistical notions of uncertainty. Based on our findings, we discuss how ADS may be better designed to support the roles of human decision-makers in social decision-making contexts.

Machine learning tools have been deployed in various contexts to support human decision-making, in the hope that human-algorithm collaboration can improve decision quality. However, the question of whether such collaborations reduce or exacerbate biases in decision-making remains underexplored. In this work, we conducted a mixed-methods study, analyzing child welfare call screen workers’ decision-making over a span of four years, and interviewing them on how they incorporate algorithmic predictions into their decision-making process. Our data analysis shows that, compared to the algorithm alone, workers reduced the disparity in screen-in rate between Black and white children from 20% to 9%. Our qualitative data show that workers achieved this by making holistic risk assessments and adjusting for the algorithm’s limitations. Our analyses also show more nuanced results about how human-algorithm collaboration affects prediction accuracy, and how to measure these effects. These results shed light on potential mechanisms for improving human-algorithm collaboration in high-risk decision-making contexts.

AI-based decision support tools (ADS) are increasingly used to augment human decision-making in high-stakes, social contexts. As public sector agencies begin to adopt ADS, it is critical that we understand workers’ experiences with these systems in practice. In this paper, we present findings from a series of interviews and contextual inquiries at a child welfare agency, to understand how they currently make AI-assisted child maltreatment screening decisions. Overall, we observe how workers’ reliance upon the ADS is guided by (1) their knowledge of rich, contextual information beyond what the AI model captures, (2) their beliefs about the ADS’s capabilities and limitations relative to their own, (3) organizational pressures and incentives around the use of the ADS, and (4) awareness of misalignments between algorithmic predictions and their own decision-making objectives. Drawing upon these findings, we discuss design implications towards supporting more effective human-AI decision-making.

Modern machine learning techniques commonly rely on complex, high-dimensional embedding representations to capture underlying structure in the data and improve performance. In order to characterize model flaws and choose a desirable representation, model builders often need to compare across multiple embedding spaces, a challenging analytical task supported by few existing tools. We first interviewed nine embedding experts in a variety of fields to characterize the diverse challenges they face and techniques they use when analyzing embedding spaces. Informed by these perspectives, we developed a novel system called Emblaze that integrates embedding space comparison within a computational notebook environment. Emblaze uses an animated, interactive scatter plot with a novel Star Trail augmentation to enable visual comparison. It also employs novel neighborhood analysis and clustering procedures to dynamically suggest groups of points with interesting changes between spaces. Through a series of case studies with ML experts, we demonstrate how interactive comparison with Emblaze can help gain new insights into embedding space structure.

Data science is the practice of deriving insight from data, enabled by modeling, computational methods, interactive visual analysis, and domain-driven problem solving. Data science draws from methodology developed in such fields as applied mathematics, statistics, machine learning, data mining, data management, visualization, and HCI. It drives discoveries in business, economy, biology, medicine, environmental science, the physical sciences, the humanities and social sciences, and beyond. Machine learning and data mining and visualization are integral parts of data science, and essential to enable sophisticated analysis of data. Nevertheless, both research areas are currently still rather separated and investigated by different communities rather independently. The goal of this workshop is to bring researchers from both communities together in order to discuss common interests, to talk about practical issues in application-related projects, and to identify open research problems. This summary gives a brief overview of the ACM KDD Workshop on Visualization in Data Science (VDS at ACM KDD and IEEE VIS), which will take place virtually on Aug 14-18, 2022 (Held in conjunction with KDD'22). The workshop website is available at http://www.visualdatascience.org/2022/

Abstract Existing visualization recommendation systems commonly rely on a single snapshot of a dataset to suggest visualizations to users. However, exploratory data analysis involves a series of related interactions with a dataset over time rather than one‐off analytical steps. We present Solas, a tool that tracks the history of a user's data analysis, models their interest in each column, and uses this information to provide visualization recommendations, all within the user's native analytical environment. Recommending with analysis history improves visualizations in three primary ways: task‐specific visualizations use the provenance of data to provide sensible encodings for common analysis functions, aggregated history is used to rank visualizations by our model of a user's interest in each column, and column data types are inferred based on applied operations. We present a usage scenario and a user evaluation demonstrating how leveraging analysis history improves in situ visualization recommendations on real‐world analysis tasks.

Data science is the practice of deriving insight from data, enabled by modeling, computational methods, interactive visual analysis, and domain-driven problem solving. Data science draws from methodology developed in such fields as applied mathematics, statistics, machine learning, data mining, data management, visualization, and HCI. It drives discoveries in business, economy, biology, medicine, environmental science, the physical sciences, the humanities and social sciences, and beyond. Machine learning and data mining and visualization are integral parts of data science, and essential to enable sophisticated analysis of data. Nevertheless, both research areas are currently still rather separated and investigated by different communities rather independently. The goal of this workshop is to bring researchers from both communities together in order to discuss common interests, to talk about practical issues in application-related projects, and to identify open research problems. This summary gives a brief overview of the ACM KDD Workshop on Visualization in Data Science (VDS at ACM KDD and IEEE VIS), which will take place virtually on Aug 14-18, 2021 (Held in conjunction with KDD'21). The workshop website is available at: http://www.visualdatascience.org/2021/

AI systems can fail to learn important behaviors, leading to real-world issues like safety concerns and biases. Discovering these systematic failures often requires significant developer attention, from hypothesizing potential edge cases to collecting evidence and validating patterns. To scale and streamline this process, we introduce crowdsourced failure reports, end-user descriptions of how or why a model failed, and show how developers can use them to detect AI errors. We also design and implement Deblinder, a visual analytics system for synthesizing failure reports that developers can use to discover and validate systematic failures. In semi-structured interviews and think-aloud studies with 10 AI practitioners, we explore the affordances of the Deblinder system and the applicability of failure reports in real-world settings. Lastly, we show how collecting additional data from the groups identified by developers can improve model performance.

Most of the work on interpretable machine learning has focused on designing either inherently interpretable models, which typically trade-off accuracy for interpretability, or post-hoc explanation systems, whose explanation quality can be unpredictable. Our method, ExpO, is a hybridization of these approaches that regularizes a model for explanation quality at training time. Importantly, these regularizers are differentiable, model agnostic, and require no domain knowledge to define. We demonstrate that post-hoc explanations for ExpO-regularized models have better explanation quality, as measured by the common fidelity and stability metrics. We verify that improving these metrics leads to significantly more useful explanations with a user study on a realistic task.

Ensuring effective public understanding of algorithmic decisions that are powered by machine learning techniques has become an urgent task with the increasing deployment of AI systems into our society. In this work, we present a concrete step toward this goal by redesigning confusion matrices for binary classification to support non-experts in understanding the performance of machine learning models. Through interviews (n=7) and a survey (n=102), we mapped out two major sets of challenges lay people have in understanding standard confusion matrices: the general terminologies and the matrix design. We further identified three sub-challenges regarding the matrix design, namely, confusion about the direction of reading the data, layered relations and quantities involved. We then conducted an online experiment with 483 participants to evaluate how effective a series of alternative representations target each of those challenges in the context of an algorithm for making recidivism predictions. We developed three levels of questions to evaluate users' objective understanding. We assessed the effectiveness of our alternatives for accuracy in answering those questions, completion time, and subjective understanding. Our results suggest that (1) only by contextualizing terminologies can we significantly improve users' understanding and (2) flow charts, which help point out the direction of reading the data, were most useful in improving objective understanding. Our findings set the stage for developing more intuitive and generally understandable representations of the performance of machine learning models.

The performance of deep learning models is dependent on the precise configuration of many layers and parameters. However, there are currently few systematic guidelines for how to configure a successful model. This means model builders often have to experiment with different configurations by manually programming different architectures (which is tedious and time consuming) or rely on purely automated approaches to generate and train the architectures (which is expensive). In this paper, we present Rapid Exploration of Model Architectures and Parameters, or REMAP, a visual analytics tool that allows a model builder to discover a deep learning model quickly via exploration and rapid experimentation of neural network architectures. In REMAP, the user explores the large and complex parameter space for neural network architectures using a combination of global inspection and local experimentation. Through a visual overview of a set of models, the user identifies interesting clusters of architectures. Based on their findings, the user can run ablation and variation experiments to identify the effects of adding, removing, or replacing layers in a given architecture and generate new models accordingly. They can also handcraft new models using a simple graphical interface. As a result, a model builder can build deep learning models quickly, efficiently, and without manual programming. We inform the design of REMAP through a design study with four deep learning model builders. Through a use case, we demonstrate that REMAP allows users to discover performant neural network architectures efficiently using visual exploration and user-defined semi-automated searches through the model space.

Whether figuring out where to eat in an unfamiliar city or deciding which apartment to live in, consumer generated data (i.e. reviews and forum posts) are often an important influence in online decision making. To make sense of these rich repositories of diverse opinions, searchers need to sift through a large number of reviews to characterize each item based on aspects that they care about. We introduce a novel system, SearchLens, where searchers build up a collection of "Lenses" that reflect their different latent interests, and compose the Lenses to find relevant items across different contexts. Based on the Lenses, SearchLens generates personalized interfaces with visual explanations that promotes transparency and enables deeper exploration. While prior work found searchers may not wish to put in effort specifying their goals without immediate and sufficient benefits, results from a controlled lab study suggest that our approach incentivized participants to express their interests more richly than in a baseline condition, and a field study showed that participants found benefits in SearchLens while conducting their own tasks.

Neural sequence-to-sequence models have proven to be accurate and robust for many sequence prediction tasks, and have become the standard approach for automatic translation of text. The models work with a five-stage blackbox pipeline that begins with encoding a source sequence to a vector space and then decoding out to a new target sequence. This process is now standard, but like many deep learning methods remains quite difficult to understand or debug. In this work, we present a visual analysis tool that allows interaction and "what if"-style exploration of trained sequence-to-sequence models through each stage of the translation process. The aim is to identify which patterns have been learned, to detect model errors, and to probe the model with counterfactual scenario. We demonstrate the utility of our tool through several real-world sequence-to-sequence use cases on large-scale models.

Clustering, the process of grouping together similar items into distinct partitions, is a common type of unsupervised machine learning that can be useful for summarizing and aggregating complex multi-dimensional data. However, data can be clustered in many ways, and there exist a large body of algorithms designed to reveal different patterns. While having access to a wide variety of algorithms is helpful, in practice, it is quite difficult for data scientists to choose and parameterize algorithms to get the clustering results relevant for their dataset and analytical tasks. To alleviate this problem, we built Clustervision, a visual analytics tool that helps ensure data scientists find the right clustering among the large amount of techniques and parameters available. Our system clusters data using a variety of clustering techniques and parameters and then ranks clustering results utilizing five quality metrics. In addition, users can guide the system to produce more relevant results by providing task-relevant constraints on the data. Our visual user interface allows users to find high quality clustering results, explore the clusters using several coordinated visualization techniques, and select the cluster result that best suits their task. We demonstrate this novel approach using a case study with a team of researchers in the medical domain and showcase that our system empowers users to choose an effective representation of their complex data.

The growing volume and variety of data presents both opportunities and challenges for visual analytics. Addressing these challenges is needed for big data to provide valuable insights and novel solutions for business, security, social media, and healthcare. In the case of temporal event sequence analytics it is the number of events in the data and variety of temporal sequence patterns that challenges users of visual analytic tools. This paper describes 15 strategies for sharpening analytic focus that analysts can use to reduce the data volume and pattern variety. Four groups of strategies are proposed: (1) extraction strategies, (2) temporal folding, (3) pattern simplification strategies, and (4) iterative strategies. For each strategy, we provide examples of the use and impact of this strategy on volume and/or variety. Examples are selected from 20 case studies gathered from either our own work, the literature, or based on email interviews with individuals who conducted the analyses and developers who observed analysts using the tools. Finally, we discuss how these strategies might be combined and report on the feedback from 10 senior event sequence analysts.

Understanding predictive models, in terms of interpreting and identifying actionable insights, is a challenging task. Often the importance of a feature in a model is only a rough estimate condensed into one number. However, our research goes beyond these naïve estimates through the design and implementation of an interactive visual analytics system, Prospector. By providing interactive partial dependence diagnostics, data scientists can understand how features affect the prediction overall. In addition, our support for localized inspection allows data scientists to understand how and why specific datapoints are predicted as they are, as well as support for tweaking feature values and seeing how the prediction responds. Our system is then evaluated using a case study involving a team of data scientists improving predictive models for detecting the onset of diabetes from electronic medical records.

Many researchers across diverse disciplines aim to analyze the behavior of cohorts whose behaviors are recorded in large event databases. However, extracting cohorts from databases is a difficult yet important step, often overlooked in many analytical solutions. This is especially true when researchers wish to restrict their cohorts to exhibit a particular temporal pattern of interest. In order to fill this gap, we designed COQUITO, a visual interface that assists users defining cohorts with temporal constraints. COQUITO was designed to be comprehensible to domain experts with no preknowledge of database queries and also to encourage exploration. We then demonstrate the utility of COQUITO via two case studies, involving medical and social media researchers.

A Data Scientist typically performs a number of tedious and time-consuming steps to derive insight from a raw data set. The process usually starts with data ingestion, cleaning, and transformation (e.g. outlier removal, missing value imputation), then proceeds to model building, and finally a presentation of predictions that align with the end-users objectives and preferences. It is a long, complex, and sometimes artful process requiring substantial time and effort, especially because of the combinatorial explosion in choices of algorithms (and platforms), their parameters, and their compositions. Tools that can help automate steps in this process have the potential to accelerate the time-to-delivery of useful results, expand the reach of data science to non-experts, and offer a more systematic exploration of the available options. This work presents a step towards this goal.

Extracting insights from temporal event sequences is an important challenge. In particular, mining frequent patterns from event sequences is a desired capability for many domains. However, most techniques for mining frequent patterns are ineffective for real-world data that may be low-resolution, concurrent, or feature many types of events, or the algorithms may produce results too complex to interpret. To address these challenges, we propose Frequence, an intelligent user interface that integrates data mining and visualization in an interactive hierarchical information exploration system for finding frequent patterns from longitudinal event sequences. Frequence features a novel frequent sequence mining algorithm to handle multiple levels-of-detail, temporal context, concurrency, and outcome analysis. Frequence also features a visual interface designed to support insights, and support exploration of patterns of the level-of-detail relevant to users. Frequence's effectiveness is demonstrated with two use cases: medical research mining event sequences from clinical records to understand the progression of a disease, and social network research using frequent sequences from Foursquare to understand the mobility of people in an urban environment.

OBJECTIVE: Common chronic diseases such as hypertension are costly and difficult to manage. Our ultimate goal is to use data from electronic health records to predict the risk and timing of deterioration in hypertension control. Towards this goal, this work predicts the transition points at which hypertension is brought into, as well as pushed out of, control. METHOD: In a cohort of 1294 patients with hypertension enrolled in a chronic disease management program at the Vanderbilt University Medical Center, patients are modeled as an array of features derived from the clinical domain over time, which are distilled into a core set using an information gain criteria regarding their predictive performance. A model for transition point prediction was then computed using a random forest classifier. RESULTS: The most predictive features for transitions in hypertension control status included hypertension assessment patterns, comorbid diagnoses, procedures and medication history. The final random forest model achieved a c-statistic of 0.836 (95% CI 0.830 to 0.842) and an accuracy of 0.773 (95% CI 0.766 to 0.780). CONCLUSIONS: This study achieved accurate prediction of transition points of hypertension control status, an important first step in the long-term goal of developing personalized hypertension management plans.

Predictive modeling techniques are increasingly being used by data scientists to understand the probability of predicted outcomes. However, for data that is high-dimensional, a critical step in predictive modeling is determining which features should be included in the models. Feature selection algorithms are often used to remove non-informative features from models. However, there are many different classes of feature selection algorithms. Deciding which one to use is problematic as the algorithmic output is often not amenable to user interpretation. This limits the ability for users to utilize their domain expertise during the modeling process. To improve on this limitation, we developed INFUSE, a novel visual analytics system designed to help analysts understand how predictive features are being ranked across feature selection algorithms, cross-validation folds, and classifiers. We demonstrate how our system can lead to important insights in a case study involving clinical researchers predicting patient outcomes from electronic medical records.

As datasets grow and analytic algorithms become more complex, the typical workflow of analysts launching an analytic, waiting for it to complete, inspecting the results, and then re-Iaunching the computation with adjusted parameters is not realistic for many real-world tasks. This paper presents an alternative workflow, progressive visual analytics, which enables an analyst to inspect partial results of an algorithm as they become available and interact with the algorithm to prioritize subspaces of interest. Progressive visual analytics depends on adapting analytical algorithms to produce meaningful partial results and enable analyst intervention without sacrificing computational speed. The paradigm also depends on adapting information visualization techniques to incorporate the constantly refining results without overwhelming analysts and provide interactions to support an analyst directing the analytic. The contributions of this paper include: a description of the progressive visual analytics paradigm; design goals for both the algorithms and visualizations in progressive visual analytics systems; an example progressive visual analytics system (Progressive Insights) for analyzing common patterns in a collection of event sequences; and an evaluation of Progressive Insights and the progressive visual analytics paradigm by clinical researchers analyzing electronic medical records.

As people continue to author and share increasing amounts of information in social media, the opportunity to leverage such information for relationship discovery tasks increases. In this paper, we describe a set of systems that mine, aggregate, and infer a social graph from social media inside an enterprise, resulting in over 73 million relationships between 450,000 people. We then describe SaNDVis, a novel visual analytics tool that supports people-centric tasks like expertise location, team building, and team coordination in the enterprise. We provide details of a 22-month-long, large-scale deployment to over 2,300 users from which we analyze longitudinal usage patterns, classify types of visual analytics queries and users, and extract dominant use cases from log and interview data. By integrating social position, evidence, and facets into SaNDVis, we demonstrate how users can use a visual analytics tool to reflect on existing relationships as well as build new relationships in an enterprise setting.

There is a growing body of research into the adoption and use of social software in enterprises. However, less is known about how groups, such as communities, use and appropriate these technologies, and the implications for community structures. In a study of 188 very active online enterprise communities, we found systematic differences in size, demographics and participation, aligned with differences in community types. Different types of communities differed in their appropriation of social software tools to create and use shared resources, and build relationships. We propose implications for design of community support features, services for potential community members, and organizations looking to derive value from online groups.

Despite the tremendous popularity of social network sites both on the web and within enterprises, the relationship information they contain may be often incomplete or outdated. We suggest a novel crowdsourcing approach that uses a game to help enrich and expand the social network topology. The game prompts players to provide the names of people who have a relationship with individuals they know. The game was deployed for a one-month period within a large global organization. We provide an analysis of the data collected through this deployment, in comparison with the data from the organization's social network site. Our results indicate that the game rapidly collects large volumes of valid information that can be used to enrich and reinforce an existing social network site's data. We point out other aspects and benefits of using a crowdsourcing game to harvest social network information.

Recent studies on people recommendation have focused on suggesting people the user already knows. In this work, we use social media behavioral data to recommend people the user is not likely to know, but nonetheless may be interested in. Our evaluation is based on an extensive user study with 516 participants within a large enterprise and includes both quantitative and qualitative results. We found that many employees valued the recommendations, even if only one or two of nine recommendations were interesting strangers. Based on these results, we discuss potential deployment routes and design implications for a stranger recommendation feature.

In this work we examine nine different sources for user similarity as reflected by activity in social media applications. We suggest a classification of these sources into three categories: people, things, and places. Lists of similar people returned by the nine sources are found to be highly different from each other as well as from the list of people the user is familiar with, suggesting that aggregation of sources may be valuable. Evaluation of the sources and their aggregates points at their usefulness across different scenarios, such as information discovery and expertise location, and also highlights sources and aggregates that are particularly valuable for inferring user similarity.

A common goal in graph visualization research is the design of novel techniques for displaying an overview of an entire graph. However, there are many situations where such an overview is not relevant or practical for users, as analyzing the global structure may not be related to the main task of the users that have semi-specific information needs. Furthermore, users accessing large graph databases through an online connection or users running on less powerful (mobile) hardware simply do not have the resources needed to compute these overviews. In this paper, we advocate an interaction model that allows users to remotely browse the immediate context graph around a specific node of interest. We show how Furnas' original degree of interest function can be adapted from trees to graphs and how we can use this metric to extract useful contextual subgraphs, control the complexity of the generated visualization and direct users to interesting datapoints in the context. We demonstrate the effectiveness of our approach with an exploration of a dense online database containing over 3 million legal citations.

Although both statistical methods and visualizations have been used by network analysts, exploratory data analysis remains a challenge. We propose that a tight integration of these technologies in an interactive exploratory tool could dramatically speed insight development. To test the power of this integrated approach, we created a novel social network analysis tool, SocialAction, and conducted four long-term case studies with domain experts, each working on unique data sets with unique problems. The structured replicated case studies show that the integrated approach in SocialAction led to significant discoveries by a political analyst, a bibliometrician, a healthcare consultant, and a counter-terrorism researcher. Our contributions demonstrate that the tight integration of statistics and visualizations improves exploratory data analysis, and that our evaluation methodology for long-term case studies captures the research strategies of data analysts.

During exploratory data analysis, visualizations are often useful for making sense of complex data sets. However, as data sets increase in size and complexity, static information visualizations decrease in comprehensibility. Interactive techniques can yield valuable discoveries, but current data analysis tools typically support only opportunistic exploration that may be inefficient and incomplete.

Over time, many people accumulate extensive email repositories that contain detailed information about their personal communication patterns and relationships. We present three visualizations that capture hierarchical, correlational, and temporal patterns present in user's email repositories. These patterns are difficult to discover using traditional interfaces and are valuable for navigation and reflection on social relationships and communication history. We interviewed users with diverse email habits and found that they were able to interpret these images and could find interesting features that were not evident to them through their standard email interfaces. The images also capture a wide range of variation in email practices. These results suggest that information visualizations of personal communications have value for end-users and analysts alike.

Abstract Due to e‐mail's ubiquitous nature, millions of users are intimate with the technology; however, most users are only familiar with managing their own e‐mail, which is an inherently different task from exploring an e‐mail archive. Historians and social scientists believe that e‐mail archives are important artifacts for understanding the individuals and communities they represent. To understand the conversations evidenced in an archive, context is needed. In this article, we present a new way to gain this necessary context: analyzing the temporal rhythms of social relationships. We provide methods for constructing meaningful rhythms from the e‐mail headers by identifying relationships and interpreting their attributes. With these visualization techniques, e‐mail archive explorers can uncover insights that may have been otherwise hidden in the archive. We apply our methods to an individual's 15‐year e‐mail archive, which consists of about 45,000 messages and over 4,000 relationships.

Social network analysis (SNA) has emerged as a powerful method for understanding the importance of relationships in networks. However, interactive exploration of networks is currently challenging because: (1) it is difficult to find patterns and comprehend the structure of networks with many nodes and links, and (2) current systems are often a medley of statistical methods and overwhelming visual output which leaves many analysts uncertain about how to explore in an orderly manner. This results in exploration that is largely opportunistic. Our contributions are techniques to help structural analysts understand social networks more effectively. We present SocialAction, a system that uses attribute ranking and coordinated views to help users systematically examine numerous SNA measures. Users can (1) flexibly iterate through visualizations of measures to gain an overview, filter nodes, and find outliers, (2) aggregate networks using link structure, find cohesive subgroups, and focus on communities of interest, and (3) untangle networks by viewing different link types separately, or find patterns across different link types using a matrix overview. For each operation, a stable node layout is maintained in the network visualization so users can make comparisons. SocialAction offers analysts a strategy beyond opportunism, as it provides systematic, yet flexible, techniques for exploring social networks.