Jean-Daniel Fekete

Kostiantyn Kucher, Nicole Sultanum, Angel Daza et al.

Appropriate evaluation and experimental design are fundamental for empirical sciences, particularly in data-driven fields. Due to the successes in computational modeling of languages, for instance, research outcomes are having an increasingly immediate impact on end users. As the gap in adoption by end users decreases, the need increases to ensure that tools and models developed by the research communities and practitioners are reliable, trustworthy, and supportive of the users in their goals. In this position paper, we focus on the issues of evaluating visual text analytics approaches. We take an interdisciplinary perspective from the visualization and natural language processing communities, as we argue that the design and validation of visual text analytics include concerns beyond computational or visual/interactive methods on their own. We identify four key groups of challenges for evaluating visual text analytics approaches (data ambiguity, experimental design, user trust, and "big picture" concerns) and provide suggestions for research opportunities from an interdisciplinary perspective.

Pantea Haghighatkhah, Mennatallah El-Assady, Jean-Daniel Fekete et al.

Current visual text analysis approaches rely on sophisticated processing pipelines. Each step of such a pipeline potentially amplifies any uncertainties from the previous step. To ensure the comprehensibility and interoperability of the results, it is of paramount importance to clearly communicate the uncertainty not only of the output but also within the pipeline. In this paper, we characterize the sources of uncertainty along the visual text analysis pipeline. Within its three phases of labeling, modeling, and analysis, we identify six sources, discuss the type of uncertainty they create, and how they propagate.

Jean-Daniel Fekete, Yifan Hu, Dominik Moritz et al.

The rapid advancement of AI is transforming human-centered systems, with profound implications for human-AI interaction, human-data interaction, and visual analytics. In the AI era, data analysis increasingly involves large-scale, heterogeneous, and multimodal data that is predominantly unstructured, as well as foundation models such as LLMs and VLMs, which introduce additional uncertainty into analytical processes. These shifts expose persistent challenges for human-data interactive systems, including perceptually misaligned latency, scalability constraints, limitations of existing interaction and exploration paradigms, and growing uncertainty regarding the reliability and interpretability of AI-generated insights. Responding to these challenges requires moving beyond conventional efficiency and scalability metrics, redefining the roles of humans and machines in analytical workflows, and incorporating cognitive, perceptual, and design principles into every level of the human-data interaction stack. This paper investigates the challenges introduced by recent advances in AI and examines how these developments are reshaping the ways users engage with data, while outlining limitations and open research directions for building human-centered AI systems for interactive data analysis in the AI era.

Marie Destandau, Jean-Daniel Fekete

Knowledge Graphs (KG) allow to merge and connect heterogeneous data despite their differences; they are incomplete by design. Yet, KG data producers need to ensure the best level of completeness, as far as possible. The difficulty is that they have no means to distinguish cases where incomplete entities could and should be fixed. We present a new visualisation tool: The Missing Path, to support them in identifying coherent subsets of entities that can be repaired. It relies on a map, grouping entities according to their incomplete profile. The map is coordinated with histograms and stacked charts to support interactive exploration and analysis; the summary of a subset can be compared with the one of the full collection to reveal its distinctive features. We conduct an iterative design process and evaluation with 9 Wikidata contributors. Participants gain insights and find various strategies to identify coherent subsets to be fixed.

Alexis Pister, Paolo Buono, Jean-Daniel Fekete et al.

We propose a new approach -- called PK-clustering -- to help social scientists create meaningful clusters in social networks. Many clustering algorithms exist but most social scientists find them difficult to understand, and tools do not provide any guidance to choose algorithms, or to evaluate results taking into account the prior knowledge of the scientists. Our work introduces a new clustering approach and a visual analytics user interface that address this issue. It is based on a process that 1) captures the prior knowledge of the scientists as a set of incomplete clusters, 2) runs multiple clustering algorithms (similarly to clustering ensemble methods), 3) visualizes the results of all the algorithms ranked and summarized by how well each algorithm matches the prior knowledge, 4) evaluates the consensus between user-selected algorithms, and 5) allows users to review details and iteratively update the acquired knowledge. We describe our approach using an initial functional prototype, then provide two examples of use and early feedback from social scientists. We believe our clustering approach offers a novel constructive method to iteratively build knowledge while avoiding being overly influenced by the results of often randomly selected black-box clustering algorithms.

Marie Destandau, Olivier Corby, Jean-Daniel Fekete et al.

Knowledge Graphs have become a ubiquitous technology powering search engines, recommender systems, connected objects, corporate knowledge management and Open Data. They rely on small units of information named triples that can be combined to form higher level statements across datasets following information needs. But data producers face a problem: reconstituting chains of triples has a high cognitive cost, which hinders them from gaining meaningful overviews of their own datasets. We introduce path outlines: conceptual objects characterizing sequences of triples with descriptive statistics. We interview 11 data producers to evaluate their interest. We present Path Outlines, a tool to browse path-based summaries, based on coordinated views with 2 novel visualisations. We compare Path Outlines with the current baseline technique in an experiment with 36 participants. We show that it is 3 times faster, leads to better task completion, less errors, that participants prefer it, and find tasks easier with it.

Christoph Kinkeldey, Jean-Daniel Fekete, Tanja Blascheck et al.

We present BitConduite, a visual analytics tool for explorative analysis of financial activity within the Bitcoin network. Bitcoin is the largest cryptocurrency worldwide and a phenomenon that challenges the underpinnings of traditional financial systems - its users can send money pseudo-anonymously while circumventing traditional banking systems. Yet, despite the fact that all financial transactions in Bitcoin are available in an openly accessible online ledger - the blockchain - not much is known about how different types of actors in the network (we call them entities) actually use Bitcoin. BitConduite offers an entity-centered view on transactions, making the data accessible to non-technical experts through a guided workflow for classification of entities according to several activity metrics. Other novelties are the possibility to cluster entities by similarity and exploration of transaction data at different scales, from large groups of entities down to a single entity and the associated transactions. Two use cases illustrate the workflow of the system and its analytic power. We report on feedback regarding the approach and the the software tool gathered during a workshop with domain experts, and we discuss the potential of the approach based on our findings.

Cagatay Turkay, Nicola Pezzotti, Carsten Binnig et al.

Data science requires time-consuming iterative manual activities. In particular, activities such as data selection, preprocessing, transformation, and mining, highly depend on iterative trial-and-error processes that could be sped-up significantly by providing quick feedback on the impact of changes. The idea of progressive data science is to compute the results of changes in a progressive manner, returning a first approximation of results quickly and allow iterative refinements until converging to a final result. Enabling the user to interact with the intermediate results allows an early detection of erroneous or suboptimal choices, the guided definition of modifications to the pipeline and their quick assessment. In this paper, we discuss the progressiveness challenges arising in different steps of the data science pipeline. We describe how changes in each step of the pipeline impact the subsequent steps and outline why progressive data science will help to make the process more effective. Computing progressive approximations of outcomes resulting from changes creates numerous research challenges, especially if the changes are made in the early steps of the pipeline. We discuss these challenges and outline first steps towards progressiveness, which, we argue, will ultimately help to significantly speed-up the overall data science process.

Nicolas Heulot, Jean-Daniel Fekete, Michael Aupetit

Multidimensional scaling allows visualizing high-dimensional data as 2D maps with the premise that insights in 2D reveal valid information in high-dimensions. However, the resulting projections suffer from artifacts such as bad local neighborhood preservation and clusters tearing. Interactively coloring the projection according to the discrepancy between original proximities relative to a reference item reveals these artifacts, but it is not clear if conveying these proximities using color and displaying only local information really helps the visual analysis of projections. We conducted a controlled experiment to investigate the relevance of this interactive technique to help the visual analysis of any projection regardless its quality. We compared the bare projection to the interactive coloring of the original proximities on different visual analysis tasks involving outliers and clusters. Results indicate that the interactive coloring is worthwhile for local tasks as it is significantly robust to projection artifacts whereas the projection is not. However this interactive technique does not help significantly for visual clustering tasks for that projections already give a suitable overview.

Jean-Daniel Fekete, Romain Primet

Exploring data requires a fast feedback loop from the analyst to the system, with a latency below about 10 seconds because of human cognitive limitations. When data becomes large or analysis becomes complex, sequential computations can no longer be completed in a few seconds and data exploration is severely hampered. This article describes a novel computation paradigm called Progressive Computation for Data Analysis or more concisely Progressive Analytics, that brings at the programming language level a low-latency guarantee by performing computations in a progressive fashion. Moving this progressive computation at the language level relieves the programmer of exploratory data analysis systems from implementing the whole analytics pipeline in a progressive way from scratch, streamlining the implementation of scalable exploratory data analysis systems. This article describes the new paradigm through a prototype implementation called ProgressiVis, and explains the requirements it implies through examples.