14.08.2023 | Fünf Beiträge auf der AMCIS 2023
Die Professur für Anwendungssysteme und E-Business ist mit fünf Beiträgen auf der American Conference on Information Systems 2023 (AMCIS 2023) vertreten:
Philipp Hartmann präsentiert den Beitrag: Explain AI-Based Essay Scorings without XAI - Empirical Investigation of an User-Centered UI Design for AI-Based AES Systems
Abstract: The lack of understandability of AI-based decisions is increasingly posing trust-related and regulatory problems. This also applies to the educational sector, where AI is a central element of modern automated essay scoring (AES) systems. However, current research on explainable AI primarily focuses on complex technical approaches. These explanations usually show a lack of understandability by the actual users, who often have no knowledge of AI. Based on an experiment with 245 students at a German university, we were able to show that even the basic principles of user interface design can improve understandability and hereby trustworthiness. Thus, the use of visual elements promotes understandability even when only little information is provided. Especially when providing further AI-specific information on the scoring of AES systems, however, it must be considered that in combination with visual elements an information congruency can be observed, leading to a cognitive overload in the worst case.
Sebastian Hobert präsentiert den Beitrag: How Today’s AI Content Generators Outperform Average Novice Students in Information Systems Exams
Abstract: The public availability of sophisticated AI content generators based on large-scale language models achieved enormous media attention in recent months. AI content generators are expected to have a major impact on future information systems developments as well as private and business life. Particularly in the field of education, the question arises as to how these innovations in AI technology will impact teaching, learning, and examination. With a specific focus on examinations in the information systems discipline, we show in this research-in-progress paper, based on an experimental field study, how a today’s AI content generator performs in solving a fundamental information systems exam. Based on our preliminary results, we found that the quality varies across different exam question types. With this research project, we offer start points and initial guidance on how educators can deal with the availability of AI content generators.
Sebastian Hobert präsentiert den Beitrag: Bringing Educational Chatbots into the Field: Towards a Semi-Automated Knowledge Base Generation App for Educational Chatbots
Abstract: The overall goal of this design-oriented research study is to foster the dissemination of educational chatbots. Even though prior research has shown that educational chatbots can have positive effects on learning and teaching, we observe that the technology is currently only used rarely. Lately, technical challenges in creating educational chatbots have been reduced. Nevertheless, a major challenge remains: the availability of tailored learning materials stored in knowledge bases. Such well-design knowledge bases are currently needed for educational chatbots that are specifically designed for courses. To support educators in creating knowledge bases, we propose a semi-automated knowledge base generation process and develop a corresponding app that uses techniques from text mining. In our evaluation, we compared the results of the semi-automated approach with a manually created knowledge base from a prior study. The results indicate that the semi-automated approach is able to provide an efficient way to create a knowledge base.
Christine Jokisch präsentiert den Beitrag: The Use of Gamification to Enhance Internal Software Quality - Structured Literature Analysis
Abstract: Internal software quality is a subarea of software quality assurance and refers to the static part of a software product, e.g., adherence to programming conventions, code complexity, or the extent of documentation. Investing in internal software quality is considered a long-term investment as it leads to increased maintainability and supportability of the software. However, the lack of immediate reward for the effort invested in internal software quality can demotivate software developers. This poses a significant risk to software projects, as poor internal software quality negatively impacts the maintainability of software. Gamification is an approach that aims to influence users' behavior by satisfying their needs, resulting in increased motivation. However, for gamification to be successful, it must be comprehensively planned and understood. Gamification's impact on internal software quality has yet to be determined. Therefore, this paper aims to analyze the use and impact of gamification in internal software quality.
Christine Jokisch präsentiert den Beitrag: Bridging the Gap: An Interview Study on Challenges in Software Testing and Educational Needs
Abstract: As information systems become increasingly complex, software testing is being recognized as a key element in this process. However, organizations continue to encounter challenges in effectively implementing and executing testing procedures, highlighting the persisting difficulty arising from a lack of knowledge and awareness. Thus, it is crucial to understand the challenges related to software testing to point out educational needs in software testing. To gain a comprehensive understanding of the current challenges related to software testing, this study conducted an empirical cross-sectional interview study with 23 experts. Additionally, a literature review was conducted to investigate if there are any discrepancies or similarities between the current challenges identified in academic research and industry, emphasizing relevant educational needs in software testing. The software testing life cycle was utilized as a framework to systematically present the results. Overall, this study provides insights into the challenges of software testing and current educational needs.