Multi-Agent and Retrieval-Augmented Techniques for Data Quality: A Systematic Literature Review
Session
Information Systems
Description
In the era of ever-growing amounts of data and enterprise analytics, managing data quality by ensuring accuracy, completeness, and consistency has become a critical prerequisite for reliable insights and decision-making. Traditional techniques such as rule-based validation, anomaly detection, and data profiling remain essential foundations for identifying and addressing common data issues in analytics warehouses. However, the increasing volume, variety, and velocity of modern data environments demand more adaptive and intelligent solutions. Recent advances in artificial intelligence, including Large Language Models, Retrieval-Augmented Generation, and multi-agent systems, introduce new opportunities for automating anomaly detection, contextual validation, and data repair. These approaches combine external knowledge retrieval, and coordinated agent-based workflows to improve the detection of data anomalies and enhance data completeness and consistency. This systematic literature review focuses on synthesizing and comparing different practices and methods recommended by researchers and practitioners, with the purpose of preparing a comprehensive view of the recent modern AI-driven approaches suggested for data quality in analytics warehouses.
Keywords:
Data Quality, Analytics Warehouses, Large Language Models, RetrievalAugmented Generation (RAG), Multi-Agent Systems
Proceedings Editor
Edmond Hajrizi
ISBN
978-9951-982-41-2
Location
UBT Lipjan, Kosovo
Start Date
25-10-2025 9:00 AM
End Date
26-10-2025 6:00 PM
DOI
10.33107/ubt-ic.2025.220
Recommended Citation
Shehu, Benjamin; Shehu, Visar; and Besimi, Adrian, "Multi-Agent and Retrieval-Augmented Techniques for Data Quality: A Systematic Literature Review" (2025). UBT International Conference. 7.
https://knowledgecenter.ubt-uni.net/conference/2025UBTIC/IS/7
Multi-Agent and Retrieval-Augmented Techniques for Data Quality: A Systematic Literature Review
UBT Lipjan, Kosovo
In the era of ever-growing amounts of data and enterprise analytics, managing data quality by ensuring accuracy, completeness, and consistency has become a critical prerequisite for reliable insights and decision-making. Traditional techniques such as rule-based validation, anomaly detection, and data profiling remain essential foundations for identifying and addressing common data issues in analytics warehouses. However, the increasing volume, variety, and velocity of modern data environments demand more adaptive and intelligent solutions. Recent advances in artificial intelligence, including Large Language Models, Retrieval-Augmented Generation, and multi-agent systems, introduce new opportunities for automating anomaly detection, contextual validation, and data repair. These approaches combine external knowledge retrieval, and coordinated agent-based workflows to improve the detection of data anomalies and enhance data completeness and consistency. This systematic literature review focuses on synthesizing and comparing different practices and methods recommended by researchers and practitioners, with the purpose of preparing a comprehensive view of the recent modern AI-driven approaches suggested for data quality in analytics warehouses.