Abstract

In generative AI, systems are intricate, with numerous attributes and various data types dependent on very large datasets. This research examines the effective intrinsic connectivity within layered structures across various applications. Currently, the connections between these layers lack a clear definition within a single Generative AI framework. The diversity and complexity of unstructured data sources pose challenges for modelling Gen AI systems and integrating their data, especially as repositories grow. The goal is to create knowledge-based links between systems through information systems (IS) connections, visual analytics, and improved data management. We examine knowledge-based information systems (IS) powered by Gen AI to understand the links between various unconventional data sources and predict AI development. We present the Design Science Information System (DSIS), which integrates various IS artefacts, unifies multiple petroleum domain areas, and explores the relationships among different information systems. We also assess the usefulness of DSIS components, focusing on their application, reusability, effectiveness, and interoperability. Additionally, we design IS solutions within a generative AI-driven database management system. We model DSIS as a Generative AI framework to examine the links between Design Science, Big Data, and Gen AI-guided Information Systems. Metadata assists in visualising, interpreting, and implementing IS articulations within AI systems. We develop attribute data views to gain a deeper understanding of generative AI and support knowledge management with a practical DSIS solution. The DSIS is an innovative, knowledge-based Generative AI system that efficiently links multiple domains and systems. It serves as a knowledge-driven solution for Generative AI, offering valuable insights for IS practitioners interested in exploring its applications across different fields.

Download

Share

COinS