Classifiers are among the most widely used supervised machine learning algorithms. Many classification models exist, and choosing the right one for a given task is difficult. During model selection and debugging, data scientists need to assess classifiers' performances, evaluate their learning behavior over time, and compare different models. Typically, this analysis is based on single-number performance measures such as accuracy. A more detailed evaluation of classifiers is possible by inspecting class errors. The confusion matrix is an established way for visualizing these class errors, but it was not designed with temporal or comparative analysis in mind. More generally, established performance analysis systems do not allow a combined temporal and comparative analysis of class-level information. To address this issue, we propose ConfusionFlow, an interactive, comparative visualization tool that combines the benefits of class confusion matrices with the visualization of performance characteristics over time. ConfusionFlow is model-agnostic and can be used to compare performances for different model types, model architectures, and/or training and test datasets. We demonstrate the usefulness of ConfusionFlow in a case study on instance selection strategies in active learning. We further assess the scalability of ConfusionFlow and present a use case in the context of neural network pruning.

In this work, we propose an interactive visual approach for the exploration and formation of structural relationships in embeddings of high-dimensional data. These structural relationships, such as item sequences, associations of items with groups, and hierarchies between groups of items, are defining properties of many real-world datasets. Nevertheless, most existing methods for the visual exploration of embeddings treat these structures as second-class citizens or do not take them into account at all. In our proposed analysis workflow, users explore enriched scatterplots of the embedding, in which relationships between items and/or groups are visually highlighted. The original high-dimensional data for single items, groups of items, or differences between connected items and groups are accessible through additional summary visualizations. We carefully tailored these summary and difference visualizations to the various data types and semantic contexts. During their exploratory analysis, users can externalize their insights by setting up additional groups and relationships between items and/or groups. We demonstrate the utility and potential impact of our approach by means of two use cases and multiple examples from various domains.

Trust calibration is essential in AI-assisted decision-making. If human users understand the rationale on which an AI model has made a prediction, they can decide whether they consider this prediction reasonable. Especially in high-risk tasks such as mushroom hunting (where a wrong decision may be fatal), it is important that users make correct choices to trust or overrule the AI. Various explainable AI (XAI) methods are currently being discussed as potentially useful for facilitating understanding and subsequently calibrating user trust. So far, however, it remains unclear which approaches are most effective. In this paper, the effects of XAI methods on human AI-assisted decision-making in the high-risk task of mushroom picking were tested. For that endeavor, the effects of (i) Grad-CAM attributions, (ii) nearest-neighbor examples, and (iii) network-dissection concepts were compared in a between-subjects experiment with \(N=501\) participants representing end-users of the system. In general, nearest-neighbor examples improved decision correctness the most. However, varying effects for different task items became apparent. All explanations seemed to be particularly effective when they revealed reasons to (i) doubt a specific AI classification when the AI was wrong and (ii) trust a specific AI classification when the AI was correct. Our results suggest that well-established methods, such as Grad-CAM attribution maps, might not be as beneficial to end users as expected and that XAI techniques for use in real-world scenarios must be chosen carefully.