Big Data and Analytics: The Path to Maturity
Permanent URI for this collection
Browse
Recent Submissions
Item Responsible Data Repurposing: A Conceptual Foundation(2025-01-07) Ogunseye, Shawn; Parsons, JeffAs organizations seek to maximize the value of their data assets, repurposing data is becoming increasingly important. Drawing on a synthesis of research across information systems, computer science and related fields, this paper characterizes data repurposing in terms of three key drivers: diversity, representativeness, and context-richness. We articulate six principles for how these drivers manifest and interact to enable responsible repurposing. We then discuss the implications of these principles for data collection, curation, and governance practices. Finally, we outline an agenda for future research and cross-sector collaboration to advance the theory and practice of data repurposing. Through this work, we aim to provide conceptual foundations and practical guidance for organizations seeking to steward their data as renewable assets for long-term value creation.Item A Design Proposal for Self-Adapting Analytic Systems: Properties and Transformation(2025-01-07) Kaisler, Stephen; Money, William; Espinosa, J. Alberto; Armour, FrankThis paper continues the development of concepts, structures and transformations for developing self-adapting analytic systems. We propose five properties that such systems must have and present some examples of base structures and subsequent transformations. We suggest a software architecture based on a virtual machine which provides an analytic infrastructure on which domain analytics can be built. The structures and transformations for developing the analytic infrastructure and domain analytics are based on previous work by Greiner (1980), Lenat (1983), and Oresky, Clarkson, Lenat, and Kaisler (1989).Item Artificial Intelligence in Data Integration: A Comprehensive Framework and Tool Evaluation(2025-01-07) Schulz, Thimo; Weinreuter, Maria Madeleine; Augenstein, DominikEffective data integration is crucial for organizations to manage and utilize vast, complex datasets. This paper presents a comprehensive framework for assessing Artificial Intelligence (AI)-based data integration tools, addressing the increasing demand for innovative solutions in this domain. Derived from a literature review, the framework encompasses 12 key dimensions including automation, data handling, support, and operational factors. Validated by industry practitioners, the framework demonstrates practical relevance and applicability. We assessed the derived key factors regarding their impact on the steps along the data integration process and applied the framework to evaluate 15 mature AI-based data integration tools. Our findings reveal that these tools reach high performances across the data integration process, however mostly focusing on single steps. Thereby, this study contributes to both theoretical understanding and practical tool selection, providing a robust foundation for future research and development in AI-driven data integration.Item Modeling a Reference Architecture for Concept Drift Adaptation Systems(2025-01-07) Trat, Martin; Elstermann, Matthes; Deckers, Jana; Ovtcharova, JivkaThe democratization of artificial intelligence (AI) technology is rapidly progressing and becoming an integral functional part of systems in diverse domains. To maximize the utility of AI in such contexts, ensuring its robustness is of critical significance. The research field of concept drift adaptation (CDA) focuses on the development of strategies aimed at sustaining AI robustness. Despite their availability, such strategies remain under-utilized for AI system design in practice. It is therefore crucial to enable AI system designers of various backgrounds to implement CDA systems as well as understand their operational intricacies easily. This work analyzes the use of the formal modeling paradigm Subject Orientation as a means to describe CDA systems in a strictly consistent manner that avoids ambiguities and leverages this paradigm to design and propose a reference architecture for such systems.Item Introduction to the Minitrack on Big Data and Analytics: The Path to Maturity(2025-01-07) Armour, Frank; Money, William; Kaisler, Stephen