Vier Beiträge auf der Americas Conference on Information Systems 2021 (AMCIS 2021)

Die Professur für Anwendungssysteme und E-Business ist mit vier Beiträgen auf der Americas Conference on Information Systems 2021 (AMCIS 2021) vertreten:

Michael Groth präsentiert den Beitrag: "Using Self-Play within Deep Q Learning to improve real-time Production Scheduling"

Abstract: Current methods of generating production schedules are either time-consuming or do not meet all productivity goals. A solution to this is to facilitate machine learning to generate production schedules. The necessary data to use machine learning is provided by the increasing use of cyber-physical systems in production environments. However, the mere acquisition of mass data through cyber-physical systems does not inherently lead to improvements. The algorithms must learn from historical situations to improve future calculations. Therefore, we implement a machine learning algorithm for real-time production scheduling in this paper. For this, we identify a suitable machine learning algorithm and meta-requirements based on a design science research approach. Finally, we evaluate the created artifact in a simulation study. The results show that the implemented solution can create schedules in real-time, which are of higher quality compared to priority rules.

Philipp Hartmann präsentiert den Beitrag: "The Intention to Participate in Online Exams – The Student Perspective"

Abstract: Studying at German universities is traditionally often focused on in-class face-to-face teaching. Following the emergence of SARS-CoV-2 and the danger of an uncontrollable spread of the COVID-19 pandemic, Germany decided to implement an almost complete lockdown in March 2020, which also affected universities. While teaching was continued using video recording, there was often no alternative to face-to-face exams. To help contain the pandemic and to cope with the organizational challenges, some universities introduced online exams. In this way, part of the responsibility was delegated to the participants themselves, without considering the additional psychological burden. To assess whether online exams are a viable alternative for the future, this article examines which factors correlates with the examinee’s intention to participate in them. It was shown that mental challenges, cheating, and the perceived suitability of online exams for fair grading are the main factors for or against the use of online exams.


Abstract: Chatbots are currently widely been used in corporate application areas and scenarios. Especially for customer support or information acquisition, these systems are surveyed and applied successfully. However, particularly for business processes, like the business travel organization, where employees often encounter problems or need further assistance, chatbots promise positive outcomes. Nevertheless, this is currently not being focused by the scientific community, resulting in missing design recommendations and unachievable possible benefits. Therefore, we conducted a Design Science Research study to (1) examine this application scenario and deduce design recommendations, as well as (2) implement a software artifact, which provides preliminary results. First, we point out eight design recommendations for business travel organization chatbots. Second, we show how chatbots should be designed for this process. Hereby the chatbot supports the process completion and adapts it to the individual user. Further, the chatbot enables situational-dependent input options and provides workflows.

Tobias Nießner präsentiert den Beitrag: "Towards a taxonomy of AI-based methods in Financial Statement Analysis"

Abstract: Artificial Intelligence (AI) is becoming more popular in a wide variety of application areas in finance. It is expected that human tasks in analyzing data can be replaced by the use of AI while saving time and costs. AI-based methods can be used to support several decision problems in the context of financial statement analysis. This paper describes the iterative development process towards a taxonomy of AI-based methods in the financial statement analysis. The purpose of the taxonomy is to create a classification pattern that can serve practitioners and researchers as a foundation for future development and measurement of different methods. Therefore, we examined criteria for developing AI-based methods, while referring to the identified major use-cases in financial statement analysis within academic literature as well as practice publications. We identified six dimensions and fifteen corresponding characteristics that refer to the developing process of AI-based methods in financial statement analysis.