Search

In this study, we introduce the Conversation Progress Guide (CPG), a system designed for text-based conversational AI interactions that provides a visual interface to represent progress. Users often encounter failures when interacting with conversational AI, which can negatively affect their self-efficacy-an individual's belief in their capabilities, reducing their willingness to engage with these services. The CPG offers visual feedback on task progress, providing users with mastery experiences, a key source of self-efficacy. To evaluate the system's effectiveness, we conducted a user study assessing how the integration of the CPG influences user engagement and self-efficacy. Results demonstrate that users interacting with a conversational AI enhanced by the CPG showed significant improvements in self-efficacy measures compared to those using a conventional conversational AI., Comment: Accepted to ACM CHI2025' more...

Precise measurements from sensors are crucial, but data is usually collected from low-cost, low-tech systems, which are often inaccurate. Thus, they require further calibrations. To that end, we first identify three requirements for effective calibration under practical low-tech sensor conditions. Based on the requirements, we develop a model called TESLA, Transformer for effective sensor calibration utilizing logarithmic-binned attention. TESLA uses a high-performance deep learning model, Transformers, to calibrate and capture non-linear components. At its core, it employs logarithmic binning to minimize attention complexity. TESLA achieves consistent real-time calibration, even with longer sequences and finer-grained time series in hardware-constrained systems. Experiments show that TESLA outperforms existing novel deep learning and newly crafted linear models in accuracy, calibration speed, and energy efficiency., Comment: Accepted by AAAI 2025 more...

We present drillboards, a technique for adaptive visualization dashboards consisting of a hierarchy of coordinated charts that the user can drill down to reach a desired level of detail depending on their expertise, interest, and desired effort. This functionality allows different users to personalize the same dashboard to their specific needs and expertise. The technique is based on a formal vocabulary of chart representations and rules for merging multiple charts of different types and data into single composite representations. The drillboard hierarchy is created by iteratively applying these rules starting from a baseline dashboard, with each consecutive operation yielding a new dashboard with fewer charts and progressively more abstract and simplified views. We also present an authoring tool for building drillboards and show how experts users can use to build up and deliver personalized experiences to a wide audience. Our evaluation asked three domain experts to author drillboards for their own datasets, which we then showed to casual end-users with favorable outcomes., Comment: Accepted to IEEE Transactions on Visualization and Computer Graphics as a Regular Paper more...

Interactive visualization editors empower people to author visualizations without writing code, but do not guide them in the art and craft of effective visual communication. In this paper, we explore the potential for using an off-the-shelf Large Language Model (LLM) to provide actionable and customized feedback to visualization designers. Our implementation, called VISUALIZATIONARY, showcases how ChatGPT can be used in this manner using two components: a preamble of visualization design guidelines and a suite of perceptual filters extracting salient metrics from a visualization image. We present findings from a longitudinal user study involving 13 visualization designers - 6 novices, 4 intermediate ones, and 3 experts - authoring a new visualization from scratch over the course of several days. Our results indicate that providing guidance in natural language using an LLM can aid even seasoned designers in refining their visualizations. All supplemental materials accompanying this paper are available at https://osf.io/v7hu8. more...

Human-centered AI (HCAI) puts the user in the driver's seat of so-called human-centered AI-infused tools (HCAI tools): interactive software tools that amplify, augment, empower, and enhance human performance using AI models. We discuss how interactive visualization can be a key enabling technology for creating such human-centered AI tools. To validate our approach, we first interviewed HCI, AI, and Visualization experts to define the characteristics of HCAI tools. We then present several examples of HCAI tools using visualization and use the examples to extract guidelines on how interactive visualization can support future HCAI tools. more...

The advent of low cost, accessible, and high performance augmented reality (AR) has shed light on a situated form of analytics where in-situ visualizations embedded in the real world can facilitate sensemaking based on the user's physical location. In this work, we identify prior literature in this emerging field with a focus on situated analytics. After collecting 47 relevant situated analytics systems, we classify them using a taxonomy of three dimensions: situating triggers, view situatedness, and data depiction. We then identify four archetypical patterns in our classification using an ensemble cluster analysis. We also assess the level which these systems support the sensemaking process. Finally, we discuss insights and design guidelines that we learned from our analysis., Comment: 19 pages more...

Designing a visualization is often a process of iterative refinement where the designer improves a chart over time by adding features, improving encodings, and fixing mistakes. However, effective design requires external critique and evaluation. Unfortunately, such critique is not always available on short notice and evaluation can be costly. To address this need, we present Perceptual Pat, an extensible suite of AI and computer vision techniques that forms a virtual human visual system for supporting iterative visualization design. The system analyzes snapshots of a visualization using an extensible set of filters - including gaze maps, text recognition, color analysis, etc - and generates a report summarizing the findings. The web-based Pat Design Lab provides a version tracking system that enables the designer to track improvements over time. We validate Perceptual Pat using a longitudinal qualitative study involving 4 professional visualization designers that used the tool over a few days to design a new visualization. more...

Attention networks, a deep neural network architecture inspired by humans' attention mechanism, have seen significant success in image captioning, machine translation, and many other applications. Recently, they have been further evolved into an advanced approach called multi-head self-attention networks, which can encode a set of input vectors, e.g., word vectors in a sentence, into another set of vectors. Such encoding aims at simultaneously capturing diverse syntactic and semantic features within a set, each of which corresponds to a particular attention head, forming altogether multi-head attention. Meanwhile, the increased model complexity prevents users from easily understanding and manipulating the inner workings of models. To tackle the challenges, we present a visual analytics system called SANVis, which helps users understand the behaviors and the characteristics of multi-head self-attention networks. Using a state-of-the-art self-attention model called Transformer, we demonstrate usage scenarios of SANVis in machine translation tasks. Our system is available at http://short.sanvis.org, Comment: VAST Short - IEEE VIS 2019 more...