Publications
Decoding Resilience: A Graph-based Approach for Organizational Resilience Assessment
Janzen, S., Harig, A., Gdanitz, N., Stein, H., Öksüz-Köster, N., Maass, W.
Conference Proceedings of International Conference on Conceptual Modeling (ER 2023), CEUR-WS.org, 11/2023
This paper proposes a graph-based method for assessing organizational resilience.
LLMRS: Unlocking Potentials of LLM-Based Recommender Systems for Software Purchase
John, A., Aidoo, T., Behmanush, H., Gunduz, I. B., Shrestha, H., Rahman, M. R., Maaß, W.
Workshop on Information Technology and Systems (WITS)
Recommendation systems are ubiquitous, from Spotify playlist suggestions to Amazon product suggestions. Nevertheless, depending on the methodology or the dataset, these systems typically fail to capture user preferences and generate general recommendations. Recent advancements in Large Language Models (LLM) offer promising results for analyzing user queries. However, employing these models to capture user preferences and efficiency remains an open question. In this paper, we propose LLMRS, an LLM-based zero-shot recommender system where we employ pre-trained LLM to encode user reviews into a review score and generate user-tailored recommendations. We experimented with LLMRS on a real-world dataset, the Amazon product reviews, for software purchase use cases. The results show that LLMRS outperforms the ranking-based baseline model while successfully capturing meaningful information from product reviews, thereby providing more reliable recommendations.
Recommendation systems are ubiquitous, from Spotify playlist suggestions to Amazon product suggestions. Nevertheless, depending on the methodology or the dataset, these systems typically fail to capture user preferences and generate general recommendations. Recent advancements in Large Language Models (LLM) offer promising results for analyzing user queries. However, employing these models to capture user preferences and efficiency remains an open question. In this paper, we propose LLMRS, an LLM-based zero-shot recommender system where we employ pre-trained LLM to encode user reviews into a review score and generate user-tailored recommendations. We experimented with LLMRS on a real-world dataset, the Amazon product reviews, for software purchase use cases. The results show that LLMRS outperforms the ranking-based baseline model while successfully capturing meaningful information from product reviews, thereby providing more reliable recommendations.
Modelling Metabolism Pathways using Graph Representation Learning for Fraud Detection in Sports
Rahman, M.R., Hussain, M., Piper, T., Geyer, H., Equey, T., Baume, N., Aikin, R., Maass, W.
In Proceedings of IEEE International Conference on Digital Health, (ICDH 23), July 2-8, Chicago, Illinois, USA, 2023
This paper presents a novel graph representation learning approach for detecting fraudulent behaviors in sports.
Monetary Valuation of Data in the Context of Accounting
Stein, H., Maass, W.
Daniel Trauth; Thomas Bergs; Wolfgang Prinz. The Monetization of Technical Data: Innovations from Industry and Research. Pages 103-116, ISBN 978-3-662-66509-1, Springer Berlin Heidelberg, Berlin, Heidelberg, 2023
Discusses approaches for monetary valuation of data assets in accounting.
PAIRS – Privacy-Aware, Intelligent and Resilient Crisis Management
Janzen, S., Ahiagble, A.P., Abdel Khaliq, L., Gdanitz, N., Saxena, P., Mithare, P., Skrytskyi, D., Maass, W.
In: Conference Proceedings (Vol. 2). International Conference on Conceptual Modeling (ER-2023)
PAIRS is a framework for intelligent and privacy-aware crisis management.
POWOP: Weather-based Power Outage Prediction
Gdanitz, N., Abdel Khaliq, L., Ahiagble, A.P., Janzen, S., Maass, W.
In: IntelliSys 2023. Intelligent Systems Conference, September 7-8, Amsterdam, Netherlands. SAI IntelliSys 2023
Presents a framework for predicting power outages using weather data and intelligent systems.
PRScalc, a privacy-preserving calculation of raw polygenic risk scores from direct-to-consumer genomics data
Sandoval, L., Jafri, S., Balasubramanian, J. B., Bhawsar, P., Edelson, J. L., Martins, Y., Maass, W., Chanock, S. J., Garcia-Closas, M., & Almeida, J. S.
Bioinformatics Advances, 3(1), vbad145
PRScalc introduces a privacy-preserving approach for polygenic risk score calculations.
QUASIM: Quantum Computing Enhanced Service Ecosystem for Simulation in Manufacturing
Agrawal, A., Stein, H., Xu, S., Janzen, S., Maass, W.
Conference Proceedings of International Conference on Conceptual Modeling (ER 2023), Project Exhibitions, CEUR-WS.org, 11/2023
QUASIM explores the use of quantum computing to enhance service ecosystems for manufacturing simulations.
SNOOP Method: Faithfulness of Text Summarizations for Single Nucleotide Polymorphisms
Maass, W., Agnes, C.K., Rahman, M.R., Almeida, J.S.
2nd Symposium on Human Partnership with Medical AI: Design, Operationalization, and Ethics at the Association for the Advancement of Artificial Intelligence (AAAI) Summer Symposium 2023, Singapore
SNOOP method evaluates the faithfulness of text summarizations for genetic variations.
Tackling Non-Transparency-Identification of hidden problems in component-based supply chains
Janzen, S., Baer, S., Ahiagble, A.P., Maass, W.
In Proceedings of 20th International Conference on Information Systems for Crisis Response and Management (ISCRAM). International Conference on Information Systems for Crisis Response and Management (ISCRAM), May 28-31, Omaha, Nebraska, USA, ISCRAM digital library
Presents methods for identifying hidden issues in component-based supply chains to improve transparency.
Towards Energy-Efficient Large-Scale Artificial Intelligence for Sustainable Data Centers
Dokic, D., Stein, H., Janzen, S., Maass, W.
INFORMATIK 2023. Gesellschaft für Informatik, Bonn. Umweltinformatik zwischen Nachhaltigkeit und Wandel (UINW). Berlin. 26.-30. September 2023
This paper addresses the challenges of energy-efficient AI in sustainable data centers.
Valuation of Personal Data in the Age of Data Ownership (Paper-a-thon)
Stein, H., Reich, R. H., Visinescu, L.
In: International Conference on Information Systems (ICIS) 2022. International Conference on Information Systems (ICIS-2022), December 9-14, Copenhagen, Denmark, 12/2022
Explores methods for valuing personal data amidst the growing focus on data ownership.
A Stress-Based Smart Retail Service in Shopping Environments: An Adoption Study
Öksüz-Köster, N.
In: Proceedings NeuroIS Retreat 2022. NeuroIS Retreat, June 14-16, Vienna, Austria
Investigates the adoption of a stress-based smart retail service in shopping environments.
AI Explainability: Embedding Conceptual Models
Maass, W., Castellanos, A., Tremblay, M., Lukyanenko, R., & Storey, V.C.
ICIS 2022 Proceedings. 12
Introduces methods to embed conceptual models into AI for improved explainability.
AI meets Design Science – Towards Design Methods for AI Systems Development
Janzen, S., Stein, H., Öksüz-Köster, N., Maass, W.
In: International Conference on Information Systems (ICIS) 2022 TREOs. International Conference on Information Systems (ICIS-2022), December 9-14, Copenhagen, Denmark
Proposes design methods for AI systems development, bridging AI and design science.