As part of the PRIME (Paderborn Research Colloquium on Information Management & Engineering), the Department of Information Systems welcomes Prof. Dr. Patrick Zschech from the University of Leipzig today. We are looking forward to his presentation ""Designing a Neural Question-Answering System for Times of (Information) Crises".
Abstract:
Eras of crisis are attributed to a high uncertainty that require rich information to navigate potentially harmful situations, such as natural disasters, attacks, or pandemics. The increased demand often leads to an ‘infodemic’ in which both correct and false information spread across all types of media. To overcome these issues, we designed a neural question-answering system (QAS) that assists in identifying and evaluating information. Based on Ingwersen’s cognitive model of information retrieval interaction and natural language processing techniques, we deduced design knowledge and instantiated a situational artifact. We performed multiple evaluation episodes to investigate the answer quality, the systems’ ability to adapt new knowledge, and the artifact’s performance compared to other standard systems through an experiment with over 100 users. In doing this, we contribute to the ongoing streams of designing QASs, dealing with information overload, detecting false information, as well as fighting times of crises.
Short Bio:
Patrick Zschech is a Professor of Intelligent Information Systems and Processes at the University of Leipzig. Before that, he was an Assistant Professor for Intelligent Information Systems at the Institute of Information Systems at FAU Nürnberg-Erlangen. Patrick Zschech completed his doctorate on “Data Science and Analytics in Industrial Maintenance” at the Chair of Business Informatics, esp. Business Intelligence Research, at TU Dresden in August 2020. Patrick Zschech’s research focuses on business analytics, machine learning, and artificial intelligence. In particular, he is concerned with the design, analysis, and use of intelligent information systems based on methods and technologies of advanced data processing (e.g., deep learning, computer vision, natural language processing, process mining). One of the main areas of interest for conducting and applying his research is the field of industrial manufacturing. In addition, he deals with the analysis and design of data science qualification programs and he investigates approaches for increasing acceptance of AI systems from a socio-technical perspective.
