The chemical reaction of imine groups with vapors trifluoroacetic anhydride (TFAA) was investigated in detail X‐ray photoelectron spectroscopy (XPS) for the potential application derivatization (CD) studies plasma treated surfaces. Imine were at first prepared by converting surface amine a polymer precursor using common vapor phase fluorine tagged aldehydes and ketones. originally low yield forming approx. 50%, performed under standard conditions dramatically enhanced up to 100% an own...

Understanding the recommendations of an artificial intelligence (AI) based assistant for decision-making is especially important in high-risk tasks, such as deciding whether a mushroom edible or poisonous. To foster user understanding and appropriate trust systems, we assessed effects explainable (XAI) methods educational intervention on AI-assisted behavior 2 × between subjects online experiment with N=410 participants. We developed novel use case which users go virtual hunt are tasked...

Explainable Artificial Intelligence (XAI) enables (AI) to explain its decisions. This holds the promise of making AI more understandable users, improving interaction, and establishing an adequate level trust. We tested this claim in high-risk task AI-assisted mushroom hunting, where people had decide whether a was edible or poisonous. In between-subjects experiment, 328 visitors Austrian media art festival played tablet-based hunting game while walking through highly immersive artificial...

Classifiers are among the most widely used supervised machine learning algorithms. Many classification models exist, and choosing right one for a given task is difficult. During model selection debugging, data scientists need to assess classifiers' performances, evaluate their behavior over time, compare different models. Typically, this analysis based on single-number performance measures such as accuracy. A more detailed evaluation of classifiers possible by inspecting class errors. The...

In this work, we propose an interactive visual approach for the exploration and formation of structural relationships in embeddings high-dimensional data. These relationships, such as item sequences, associations items with groups, hierarchies between groups items, are defining properties many real-world datasets. Nevertheless, most existing methods treat these structures second-class citizens or do not take them into account at all. our proposed analysis workflow, users explore enriched...

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 consider this prediction reasonable. Especially high-risk tasks such as mushroom hunting (where wrong decision may be fatal), it important that make correct choices to trust or overrule AI. Various explainable (XAI) methods are currently being discussed potentially useful for facilitating understanding and subsequently...

Aluminum-aluminum thermo-compression wafer bonding is becoming increasingly important in the production of microelectromechanical systems (MEMS) devices. As chemically highly stable aluminum oxide layer acts as a diffusion barrier between two metallization layers, up to now process has required temperatures 300°C or more. By using EVG ® 580 ComBond system, which surface treatment and subsequent are both performed high vacuum cluster, for first time successful Al-Al was possible at...

Classification is one of the most important supervised machine learning tasks. During training a classification model, instances are fed to model multiple times (during epochs) in order iteratively increase performance. The increasing complexity models has led growing demand for interpretability through visualizations. Existing approaches mostly focus on visual analysis final performance after and often limited aggregate measures. In this paper we introduce InstanceFlow, novel dual-view...

Abstract The increasing involvement of Artificial Intelligence (AI) in moral decision situations raises the possibility users attributing blame to AI-based systems for negative outcomes. In two experimental studies with a total $$N = 911$$ participants, we explored attribution and underlying reasoning. Participants had classify mushrooms pictures as edible or poisonous support an app. Afterwards, participants read fictitious scenario which misclassification due erroneous AI recommendation...

Aluminum–aluminum wafer bonding is becoming increasingly important in the production of CMOS microelectromechanical systems. So far, successful has required extreme processing temperatures 450 °C or more, because chemically highly stable oxide layer acts as a diffusion barrier between two aluminum metallization layers. By using ComBond® system, which surface treatment and subsequent are both performed high vacuum cluster, for first time Al–Al was possible at temperature 150 °C. The bonded...

Abstract Functional conductive polymers represent an emerging class of nonmetallic electrocatalysts attractive to substitute scarce and expensive elements used today. Here the synthesis emeraldine‐polyguanine as structural‐molecular analog polyaniline is shown conducting biopolymer molecular hydrogen electrocatalyst applied. The polymerized protonated nucleobase aids catalytic electroreduction protons over a peculiar mechanism, where, amino‐association, coserve dopant reactant. In acidic...

ParaDime is a framework for parametric dimensionality reduction (DR). In DR, neural networks are trained to embed high-dimensional data items in low-dimensional space while minimizing an objective function. builds on the idea that functions of several modern DR techniques result from transformed inter-item relationships. It provides common interface specifying these relations and transformations defining how they used within losses govern training process. Through this interface, unifies...

Understanding the recommendation of an artificially intelligent (AI) assistant for decision-making is especially important in high-risk tasks, such as deciding whether a mushroom edible or poisonous. To foster user understanding and appropriate trust systems, we tested effects explainable artificial intelligence (XAI) methods educational intervention on AI-assisted behavior 2x2 between subjects online experiment with N = 410 participants. We developed novel use case which users go virtual...

Abstract Dashboards are used ubiquitously to gain and present insights into data by means of interactive visualizations. To bridge the gap between non‐expert dashboard users potentially complex datasets and/or visualizations, a variety onboarding strategies employed, including videos, narration, tutorials. We propose process model for that formalizes unifies such diverse strategies. Our introduces loop alongside usage loop. Unpacking reveals how each strategy combines selected building...

Explainable Artificial Intelligence (XAI) enables an (AI) to explain its decisions. This holds the promise of making AI more understandable users, improving interaction, and establishing adequate level trust. We tested this assertion in high-risk task mushroom hunting, where users have decide whether a is edible or poisonous with aid AI-based app that suggests classifications based on images. In between-subjects experiment N = 328 visitors Austrian media art exhibition played hunting game...

Onboarding a user to visualization dashboard entails explaining its various components, including the chart types used, data loaded, and interactions available. Authoring such an onboarding experience is time-consuming requires significant knowledge little guidance on how best complete this task. Depending their levels of expertise, end users being onboarded new can be either confused overwhelmed or disinterested disengaged. We propose interactive tours (D-Tours) as semi-automated...

In this work we propose Marjorie, a visual analytics approach to address the challenge of analyzing patients' diabetes data during brief regular appointments with their diabetologists.Designed in consultation diabetologists, Marjorie uses combination and algorithmic methods support exploration patterns data.Patterns interest include seasonal variations glucose profiles, non-periodic such as fluctuations around mealtimes or periods hypoglycemia (i.e., levels below normal range).We introduce...

In this work we propose Marjorie, a visual analytics approach to address the challenge of analyzing patients' diabetes data during brief regular appointments with their diabetologists. Designed in consultation diabetologists, Marjorie uses combination and algorithmic methods support exploration patterns data. Patterns interest include seasonal variations glucose profles, non-periodic such as fuctuations around mealtimes or periods hypoglycemia (i.e., levels below normal range). We introduce...