Fraunhofer Institute for Applied Information Technology

As population aging will likely lead to an increasing number of people in need of care, the demand for informal care is expected to rise. In this context, it is often discussed whether financial incentives can motivate more individuals to assume caregiving responsibilities. We analyze the potential effect of financial incentives on the provision of informal care by estimating a structural model with endogenous labor supply and caregiving decisions. This allows us to investigate how both individual wages and financial compensations for caregiving affect the caregiving decision, while accounting for heterogeneous preferences. We find that wage increases are associated with a decreased willingness to care. Financially compensating potential carers for the opportunity costs from caregiving significantly increases the probability of providing care. However, across different subgroups, a large share of about 50% of potential carers remains unwilling to provide care despite the financial incentive. For these individuals, factors such as preferences and social norms outweigh financial considerations in their caregiving decision.

Process mining (PM) technology evolves around the analysis, design, implementation, and ongoing improvement of business processes. While it has experienced a lot of attention and significant technological advancements, contributions to the field have mostly revolved around technical matters, neglecting managerial and organizational aspects. Thus, researchers have called for a more holistic view of the application and adoption of PM in enterprises. To address this gap, this paper presents a taxonomy for organizational PM setups. Its applicability and usefulness are shown in three exemplary cases. This study extends the descriptive knowledge at the intersection of PM and business process management governance, highlighting the unique governance requirements associated with PM that cannot be effectively addressed through traditional governance approaches. The taxonomy provides practitioners with orientation when developing an effective PM setup and helps to characterize existing setups.

AI-powered systems pose unknown challenges for designers, policymakers, and users, making it more difficult to assess potential harms and outcomes. Although understanding risks is a requirement for building trust in technology, users are often excluded from risk assessments and explanations in policy and design. To address this issue, we conducted three workshops with 18 participants and discussed the EU AI Act, which is the European proposal for a legal framework for AI regulation. Based on results of these workshops, we propose a user-centered conceptual model with five risk dimensions (Design and Development, Operational, Distributive, Individual, and Societal) that includes 17 key risks. We further identify six criteria for categorizing use cases. Our conceptual model (1) contributes to responsible design discourses by connecting the risk assessment theories with user-centered approaches, and (2) supports designers and policymakers in more strongly considering a user perspective that complements their own expert views.

Contribution: This research explores the benefits and challenges of developing, deploying, and evaluating a large language models (LLMs) chatbot, MoodleBot, in computer science classroom settings. It highlights the potential of integrating LLMs into LMSs like Moodle to support self-regulated learning (SRL) and help-seeking behavior. Background: Computer science educators face immense challenges incorporating novel tools into LMSs to create a supportive and engaging learning environment. MoodleBot addresses this challenge by offering an interactive platform for both students and teachers. Research Questions: Despite issues like bias, hallucinations, and teachers’ and educators’ resistance to embracing new (AI) technologies, this research investigates two questions: (RQ1) To what extent do students accept MoodleBot as a valuable tool for learning support? (RQ2) How accurately does MoodleBot churn out responses, and how congruent are these with the established course content? Methodology: This study reviews pedagogical literature on AI-driven chatbots and adopts the retrieval-augmented generation (RAG) approach for MoodleBot’s design and data processing. The technology acceptance model (TAM) evaluates user acceptance through constructs like perceived usefulness (PU) and Ease of Use. Forty-six students participated, with 30 completing the TAM questionnaire. Findings: LLM-based chatbots like MoodleBot can significantly improve the teaching and learning process. This study revealed a high accuracy rate (88%) in providing course-related assistance. Positive responses from students attest to the efficacy and applicability of AI-driven educational tools. These findings indicate that educational chatbots are suitable for integration into courses to improve personalized learning and reduce teacher administrative burden, although improvements in automated fact-checking are needed.

In traditional financial markets, front-running is a well-structured phenomenon. It represents a form of privileged actors utilizing knowledge or power advantages to extract undue profit at the cost of other stakeholders. Various mitigation strategies have emerged, ranging from market design to regulatory measures. More recently, a similar and substantially richer variety of means to gain unethical profit from power asymmetries has appeared in the context of blockchain-based decentralized applications. This phenomenon is called “maximal extractable value” (MEV). Despite the decentralized nature and inherent transparency of blockchain ledgers, MEV is particularly prevalent and challenging to mitigate. While related work in computer science and algorithmic game theory has already identified several different ways in which MEV manifests in decentralized finance (DeFi) and outlined partial solution approaches, a discussion of its impacts in the information systems (IS) domain is still absent. A holistic definition of MEV and how it can be exploited is necessary for the discussion of its potential implications for blockchain-based IS for businesses and public institutions. This paper conducts a systematic literature review to close this gap. It consolidates the diverging definitions of MEV and provides a categorization of the different ways in which it can manifest. As such, we synthesize and review the existing state of knowledge on MEV and point to undiscovered areas relevant to decentralized electronic markets in the form of a research agenda.

The success of palliative care requires collaboration among multiple professions within a sensibly digitized work system. The diverse perspectives and expertise of team members inform their collective endeavor, often leading to differing interpretations and priorities in patient care. This diversity necessitates a continual exchange of knowledge and information. Current technologies, including the hospital information system, do not foster such collaboration, particularly in palliative care. This study explores digital enhancements that can promote multi-professional collaboration (MPC). The authors employed action design research and used a work system theory lense to examine the palliative care work systems in two hospital wards in Germany. Through extensive on-site observations and interventions with practitioners, the study identified challenges that arose during MPC. This paper presents the proposed organizational and technical solutions. The paper provides design principles and guidelines for a collaboration support system to maximize MPC. Theoretical contributions include insights into the challenges of MPC and design knowledge about collaboration support. This work can inform practitioners about common challenges and offers potential solutions and guidance for implementing a collaboration support system.

The application of uncertainty-aware visualization techniques in Machine Learning (ML) predictions has proven to be invaluable in the realm of clinical data. This article delves into the prospect of transferring these lessons to sporting applications. By scrutinizing the insights derived from uncertainty-aware visualization in clinical data, our goal is to harness the potential of these techniques and apply them to augment the analysis and interpretation of ML predictions in sports. The article underscores the significance of comprehending and visually representing uncertainty in sporting data, elucidating various visualization methods, including error bars, heatmaps, probability distributions, ensemble methods, and sensitivity analysis. Through this exploration, we illustrate how uncertainty-aware visualization can contribute to enhancing the reliability and decision-making processes associated with ML predictions in sports. Drawing upon the knowledge acquired from uncertainty-aware visualization in clinical data, we can lay the groundwork for more resilient and informed applications of ML in the sporting domain.

Charging electric vehicles (EVs) with renewable energy can lessen their environmental impact. However, the fluctuating availability of renewable energy affects the sustainability of public EV charging stations. Nearby public charging stations may utilize differing energy sources due to their microgrid connections-ranging from exclusively renewable to non-renewable or a combination of both-highlighting the substantial variability in energy supply types within short distances. This study investigates the near-future scenario of integrating dynamic renewable energy availability in charging station navigation to impact the choices of EV users towards renewable sources. We conducted a within-subjects design survey with 50 car users and semi-structured interviews with 10 EV users from rural, suburban, and urban areas. The results show that when choosing EV charging stations, drivers often prioritize either time savings or money savings based on the driving scenarios that influence drivers' consumer value. Notably, EV users tend to select renewable-powered stations when they align with their main priority , be it saving money or time. This study offers end-user insights into the front-end graphic user interface and the development of the back-end ranking algorithm for navigation recommender systems that integrate dynamic renewable energy availability for the sustainable use of electric vehicles. CCS CONCEPTS • Human-centered computing → User centered design.

Twin Transformation” is characterised by synergistic leveraging of efforts towards digital and sustainability transformation. It relies on digital transformation to develop digital solutions that can improve sustainability and on sustainability transformation to provide the goals and insights that are required to design these digital solutions. This integrated approach uses data streams and the predictive and generative capabilities of systems enabled by Artificial Intelligence (AI). These systems help to overcome the boundaries of human rationality in addressing the complex problem space that exists at the intersection of digital and sustainability transformation. This chapter provides a framework for AI-enabled Twin Transformation and calls for a joint discourse to master what are arguably the two key transformations of this and the following decades.

In the United States, there is a proposal to link hospital Medicare payments with health equity measures, signaling a need to precisely measure equity in healthcare delivery. Despite significant research demonstrating disparities in health care outcomes and access, there is a noticeable gap in tools available to assess health equity across various health conditions and treatments. The available tools often focus on a single area of patient care, such as medication delivery, but fail to examine the entire health care process. The objective of this study is to propose a process mining framework to provide a comprehensive view of health equity. Using event logs which track all actions during patient care, this method allows us to look at disparities in single and multiple treatment steps, but also in the broader strategy of treatment delivery. We have applied this framework to the management of patients with sepsis in the Intensive Care Unit (ICU), focusing on sex and English language proficiency. We found no significant differences between treatments of male and female patients. However, for patients who don’t speak English, there was a notable delay in starting their treatment, even though their illness was just as severe and subsequent treatments were similar. This framework subsumes existing individual approaches to measure health inequities and offers a comprehensive approach to pinpoint and delve into healthcare disparities, providing a valuable tool for research and policy-making aiming at more equitable healthcare.

The paper discusses biases in medical imaging analysis, particularly focusing on the challenges posed by the development of machine learning algorithms and generative models. It introduces a taxonomy of bias problems and addresses them through a data infrastructure initiative: the PADME (Platform for Analytics and Distributed Machine-Learning for Enterprises), which is a part of the National Research Data Infrastructure for Personal Health Data (NFDI4Health) project. The PADME facilitates the structuring and sharing of health data while ensuring privacy and adherence to FAIR principles. The paper presents experimental results that show that generative methods can be effective in data augmentation. Complying with PADME infrastructure, this work proposes a solution framework to deal with bias in the different data stations and preserve privacy when transferring images. It highlights the importance of standardized data infrastructure in mitigating biases and promoting FAIR, reusable, and privacy-preserving research environments in healthcare.

The governing coalition would like to translate the existing income tax deduction for single parents into a tax relief. The aim is to treat all single parents more equally, regardless of the amount of their taxable income. We use detailed microdata from official tax returns to analyze the distributional effects of both, the present tax deduction and those of a revenue-neutral tax relief. Our analysis shows that the switch from deduction to relief will benefit single parents with low incomes. This comes at the expense of higher incomes. Overall, the number of reform winners outnumbers those of the reform losers.

The German coalition agreement (2021) intends to transform withholding tax class combination III/V for married couples to the symmetric tax class combination IV/IV with a factor, to strengthen economic independence and fairness. The taxation with a factor could eliminate possible disadvantages of tax class combination III/V, particularly the high marginal tax burden in tax class V. The planned reform is associated with several pitfalls. We present the expected reform effects on tax revenues and income distribution, which we calculate using a microsimulation model. We then discuss possible difficulties associated with the reform’s implementation, in particular those concerning individuals employed less than 12 months and sole earners.

Background Pneumonia and lung cancer have a mutually reinforcing relationship. Lung cancer patients are prone to contracting COVID-19, with poorer prognoses. Additionally, COVID-19 infection can impact anticancer treatments for lung cancer patients. Developing an early diagnostic system for COVID-19 pneumonia can help improve the prognosis of lung cancer patients with COVID-19 infection. Method This study proposes a neural network for COVID-19 diagnosis based on non-enhanced CT scans, consisting of two 3D convolutional neural networks (CNN) connected in series to form two diagnostic modules. The first diagnostic module classifies COVID-19 pneumonia patients from other pneumonia patients, while the second diagnostic module distinguishes severe COVID-19 patients from ordinary COVID-19 patients. We also analyzed the correlation between the deep learning features of the two diagnostic modules and various laboratory parameters, including KL-6. Result The first diagnostic module achieved an accuracy of 0.9669 on the training set and 0.8884 on the test set, while the second diagnostic module achieved an accuracy of 0.9722 on the training set and 0.9184 on the test set. Strong correlation was observed between the deep learning parameters of the second diagnostic module and KL-6. Conclusion Our neural network can differentiate between COVID-19 pneumonia and other pneumonias on CT images, while also distinguishing between ordinary COVID-19 patients and those with white lung. Patients with white lung in COVID-19 have greater alveolar damage compared to ordinary COVID-19 patients, and our deep learning features can serve as an imaging biomarker.

Processes in real-life scenarios tend to inherently establish partial orders over their constituent activities. This makes partially ordered graphs viable for process modeling. While partial orders capture both concurrent and sequential interactions among activities in a compact way, they fall short in modeling choice and cyclic behavior. To address this gap, we introduce the Partially Ordered Workflow Language (POWL), a novel language for process modeling that combines traditional hierarchical modeling languages with partial orders. In a POWL model, sub-models are combined into larger ones either as partial orders or using control-flow operators that enable the representation of choice and loop structures. This integration of hierarchical structure and partial orders not only offers an effective solution for process modeling but also provides quality guarantees that make POWL particularly suitable for the automated discovery of process models.

ProMoAI is a novel tool that leverages Large Language Models (LLMs) to automatically generate process models from textual descriptions, incorporating advanced prompt engineering, error handling, and code generation techniques. Beyond automating the generation of complex process models, ProMoAI also supports process model optimization. Users can interact with the tool by providing feedback on the generated model, which is then used for refining the process model. ProMoAI utilizes the capabilities LLMs to offer a novel, AI-driven approach to process modeling, significantly reducing the barrier to entry for users without deep technical knowledge in process modeling.

Macro-task crowdsourcing presents a promising approach to address wicked problems like climate change by leveraging the collective efforts of a diverse crowd. Such macro-task crowdsourcing requires facilitation. However, in the facilitation process, traditionally aggregating and synthesizing text contributions from the crowd is labor-intensive, demanding expertise and time from facilitators. Recent advancements in large language models (LLMs) have demonstrated human-level performance in natural language processing. This paper proposes an abstract design for an information system, developed through four iterations of a prototype, to support the synthesis process of contributions using LLM-based natural language processing. The prototype demonstrated promising results, enhancing efficiency and effectiveness in synthesis activities for macro-task crowdsourcing facilitation. By streamlining the synthesis process, the proposed system significantly reduces the effort to synthesize content, allowing for stronger integration of synthesized content into the discussions to reach consensus, ideally leading to more meaningful outcomes.

Given cities’ rising environmental problems and increasing food insecurity, innovative organizational endeavors such as urban agriculture present a chance for additional ecosystem services and food production. However, urban spaces are hostile as they jeopardize the availability of air, water, or soil. While digital innovations enable the management of scarce resources in traditional agricultural contexts, little is known about their applicability in urban agriculture endeavors. This study proposes a multi-layer taxonomy focusing on digital technologies, data, and different approaches in urban agriculture, as well as 20 organizational readiness factors derived with academics and practitioners from the smart urban agriculture domain. Combining both perspectives, the study sheds light on the nature of smart urban agriculture and ways to leverage its economic, ecological, and social value.