In information systems (IS) research, taxonomies are an important means of structuring and categorising knowledge. They provide a basis for systematic analyses and are often further developed in science to reflect progress in research. This raises the question of the extent to which these taxonomies are continuously and cumulatively developed and how they serve as a basis for subsequent studies.

The aim of this bachelor's thesis is to systematically analyse the most cited taxonomy publications in the field of IS research from 2015 to 2019. The aim is to analyse whether and how these taxonomies were further developed in subsequent studies either by the original authors or by other researchers, whether, for example, archetypes were used, empirical correlations of new constructs were examined or other forms of further development were undertaken. The aim is to analyse which methodological approaches and strategies were used for cumulative further development and how these contribute to the advancement of knowledge in IS research. Finally, by analysing these publications, patterns in the cumulative development of taxonomies should be identified and insights gained for the structured further development of future taxonomies.

Nickerson, R. C., Varshney, U., & Muntermann, J. (2013). A method for taxonomy development and its application in information systems. European Journal of Information Systems, 22(3), 336-359.

Kundisch, D., Muntermann, J., Oberländer, A. M., Rau, D., Röglinger, M., Schoormann, T., & Szopinski, D. (2021). An update for taxonomy designers: methodological guidance from information systems research. Business & Information Systems Engineering, 1-19.

In IS research, there are now numerous business model taxonomies that allow structuring and classifying different business models. Many of these taxonomies have been developed specifically for certain industries or application areas, which often leads to overlaps in the dimensions and characteristics.

The aim of this bachelor thesis is to create a comprehensive overview of existing business model taxonomies, analyse them for similarities and differences and identify potential overlaps in the dimensions and characteristics used. In this way, a basis is to be created that helps to achieve better comparability and comprehensibility of the various taxonomies and, if necessary, to identify potential for harmonisation.

Sterk, F., Stocker, A., Heinz, D., & Weinhardt, C. (2024). Unlocking the value from car data: A taxonomy and archetypes of connected car business models. Electronic Markets, 34(1), 13.

Böttcher, T. P., Empelmann, S., Weking, J., Hein, A., & Krcmar, H. (2024). Digital sustainable business models: Using digital technology to integrate ecological sustainability into the core of business models. Information Systems Journal, 34(3), 736-761.

Schoormann, T., Schweihoff, J., Jussen, I., & Möller, F. (2023). Classification tools for business models: Status quo, comparison, and agenda. Electronic Markets, 33(1), 7.

The AI Act of the European Union and its requirements for Explainable Artificial Intelligence (XAI)

The AI Act of the European Union (EU), which was adopted last year, is the first comprehensive legal framework for the use of AI systems worldwide. The AI regulation takes a risk-based approach. Accordingly, high-risk AI systems in particular must fulfil strict requirements in terms of transparency, traceability and explainability in order to be approved for the EU market. Lower-risk applications are subject to less stringent transparency and information requirements. With this set of rules, the ability of an AI system to explain decisions in a comprehensible manner is no longer just a technical or ethical requirement, but a legal requirement. Developers and providers of AI systems must therefore be prepared to design their systems to be not only performant but also legally compliant.

Explainable Artificial Intelligence (XAI) is becoming increasingly important in the context of these developments. XAI comprises a range of methods and techniques that aim to make the behaviour and decisions of machine learning models comprehensible to humans. The use of these methods and techniques can contribute in part to compliance with the legal requirements of the AI Act. However, the definition of transparency in XAI research differs from that of the AI Act. While XAI research describes transparency as the comprehensibility and accessibility of how an algorithm works, the EU AI Act understands transparency as a means to achieve broader goals, such as accountability if the result of an AI turns out to be wrong. The legal perspective thus goes beyond technical interpretation and places new demands on the design of explainable AI systems.

The bachelor's thesis therefore aims to analyse what requirements the EU AI Act places on XAI and to what extent current research in the field of XAI meets these requirements. The starting point is a conceptual analysis of the EU AI Act, supplemented by a systematic literature review of existing XAI approaches. The first step is to identify, define and differentiate between basic terms and concepts in the EU AI Act and in the field of XAI. In a second step, the procedure of the systematic literature research and the results obtained will be documented and presented. The identified research contributions are to be differentiated from one another by means of a suitable categorisation. Based on this categorisation, it will be determined in which fields existing XAI research already meets the requirements and where there is potential for development with regard to the requirements of the AI Act. The aim of this bachelor thesis is therefore to systematically work out, based on the current state of legislation and XAI research, how the EU AI Act sets requirements for XAI and in which fields existing XAI approaches do not yet fully meet these requirements or where there is further potential for development.

Brasse, J., Broder, H. R., Förster, M., Klier, M., & Sigler, I. (2023). Explainable artificial intelligence in information systems: A review of the status quo and future research directions. Electronic Markets, 33(1), p. 26.

Gyevnar, B., Ferguson, N., & Schafer, B. (2023). Bridging the Transparency Gap: What Can Explainable AI Learn from the AI Act? ECAI 2023, pp. 964-971.

Panigutti, C., Hamon, R., Hupont, I., Fernandez Llorca, D., Fano Yela, D., Junklewitz, H., Scalzo, S., Mazzini, G., Sanchez, I., Garrido, J. S., & Gomez, E. (2023). The role of explainable AI in the context of the AI Act. Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency, pp. 1139-1150.

AI Literacy through Explainability - Teaching AI Literacy with Explainable Artificial Intelligence (XAI)

With the increasing spread of artificial intelligence (AI) in everyday life and in the world of work, there is also a growing social need to promote basic knowledge and skills in dealing with AI, so-called AI Literacy. AI literacy includes, among other things, the ability to communicate with AI systems, interact effectively and critically scrutinise their decisions.

There is new potential for teaching these skills, particularly in the context of explainable artificial intelligence (XAI), as XAI aims to make decisions made by AI systems and the way they work comprehensible to humans. By providing suitable explanations in visual, textual or interactive form, XAI can help to make complex decision-making processes more understandable and accessible and thus also promote users' technical understanding. For this reason, existing research recommends the use of XAI in the design of measures to promote learning when dealing with AI systems. However, consideration should be given to which and how many components of the AI should be explained, as explaining too many components can overwhelm users, while explaining too few components can impair learning effectiveness. The impact of explanations on learning success can also vary across user groups with different demographic characteristics. In XAI research, there are already a number of studies looking at these correlations. However, as research on AI literacy in this context is still in its early stages, little is known about the conditions under which XAI actually contributes to skill building. At this interface, the question arises as to what conditions must be met for XAI to contribute effectively to the promotion of AI literacy.

The bachelor's thesis therefore aims to systematically analyse the extent to which XAI can serve as a means of improving AI literacy. To this end, a well-founded overview of existing research in the field of XAI and AI literacy will be developed. In particular, the focus will be on studies that investigate how explanations influence the understanding and competence of users in dealing with AI applications. Firstly, key terms and concepts relating to XAI and AI literacy will be identified, defined and differentiated from one another. This will be followed by a systematic literature review, the methodological approach and the results obtained will be documented and presented in a comprehensible manner. The identified research contributions are to be structured by means of a suitable categorisation within the framework of a concept matrix. Based on this systematisation, the aim is to work out how XAI can contribute to the promotion of AI literacy and what implications can be derived from this for research and practice in the field of digital education and XAI.

Long, D., & Magerko, B. (2020). What is AI literacy? Competencies and design considerations. Proceedings of the 2020 CHI conference on human factors in computing systems, pp. 1-16.

Miller, T. (2019). Explanation in artificial intelligence: Insights from the social sciences. Artificial intelligence, 267, pp. 1-38.

Ng, D. T. K., Leung, J. K. L., Chu, S. K. W., & Qiao, M. S. (2021). Conceptualising AI literacy: An exploratory review. Computers and Education: Artificial Intelligence, 2.

Current research in the field of information systems (IS) criticises that existing taxonomies often do not use optimal cluster analysis methods. Heumann et al. (2024) emphasises this problem and argues that the choice of unsuitable methods impairs the validity and informative value of taxonomies. In order to increase the quality and accuracy of classifications in IS research, it is therefore essential to apply the proposed improved methods of cluster analysis and to investigate their effects.

The aim of this master's thesis is to replicate two existing studies (1. taxonomy on museum admission prices and 2. taxonomy on business models of cultural event platforms). In each case, the cluster analysis is to be carried out using the methodology proposed in the critical paper. The replication and comparison with the original results will analyse the extent to which the improved cluster analysis leads to more precise and meaningful taxonomies. This helps to fill methodological gaps in IS research and provides valuable insights for future studies in the field.

Good command of English

Knowledge of statistical analysis techniques and Python an advantage

Heumann, M., Kraschewski, T., Werth, O., & Breitner, M. H. (2024). Reassessing taxonomy-based data clustering: Unveiling insights and guidelines for application. Decision Support Systems, 114344.

Althaus, M., Mueller, S. J., & Kundisch, D. (2024). What price culture?-a taxonomy of the admission pricing policy at museums. International Journal of Cultural Policy, 30(3), 392-407.