Creating Data Transparency with the Support of LLMs and AI
Data Transparency is a basic need of every data worker. It is crucial to find the right data in a limited amount of time to keep the time-to-market of data and analytical products short. However, documenting and classifying data manually can be cumbersome due to the vast amount of data. In this session, we present the approach MediaMarktSaturn has taken to use LLMs and other AI models in combination with a data catalog to establish a high level of data transparency in a semi-automated way.
Target Audience: Chief Data Officers, Data Governance Managers, Data Strategists, Data Engineers, Data Scientists, Subject Matter Experts
Prerequisites: Basic knowledge of Data Catalogs and LLMs
Level: Basic
Extended Abstract:
Data Transparency is a basic need of every data worker. It is crucial to find the right data in a limited amount of time to keep the time-to-market of data and analytical products short. But with the ever-growing amount of data, it gets more and more difficult to keep up a suitable level of data transparency. In this session, we present the approach MediaMarktSaturn has taken to use LLMs and Open Source Data Catalogs to establish a high level of data transparency in a semi-automated way. With this approach, it is possible to keep the manual work at a suitable level and manage data transparency in an efficient way. Moreover, we will outline how this approach is integrated into the overall Data Governance and Data Architecture.
Dieter Berwald is Competency Lead and Product Owner at MediaMarktSaturn Technology, where he focuses on advanced analytics and data catalog applications. He currently leads projects aimed at entity recognition in large cloud data lakes, managing data access, and leveraging metadata for enhanced data utilization.
Dr. Christian Fürber ist promovierter Datenqualitätsexperte und Geschäftsführer der Information Quality Institute GmbH (iqinstitute.de), einem spezialisierten Beratungsunternehmen für Data Excellence und Data Management Lösungen. Vor der Gründung von IQI im Jahr 2012 war er in verschiedenen Positionen im Datenmanagement tätig. Unter anderem etablierter er für die Bundeswehr eine der ersten Data Governance Organisationen und führte dort ein standardisiertes Datenqualitätsmanagement ein. Christian und sein Team haben unzählige Datenprojekte und -strategien für bekannte Unternehmen verschiedener Branchen erfolgreich umgesetzt und ihnen dabei geholfen, erheblichen Mehrwert aus ihren Daten zu ziehen. Neben seiner Tätigkeit bei IQI ist Christian auch Autor und Redner auf Konferenzen (u.a. am MIT) und organisiert den TDWI Themenzirkel "Data Strategy & Data Governance".
Nitin embarked on his IT career in 2014 as a consultant, quickly gaining a reputation for his expertise and adaptability. Throughout his career, he has worked with a diverse range of clients across industries including telecommunications, petroleum, and disaster recovery. Nitin holds a Master's degree from the prestigious Technical University of Munich (TUM), which further solidified his foundation in data architecture and engineering.
Following his studies, Nitin excelled as a Lead Data Architect/Data Engineer at several renowned companies. His innovative cloud solution architecture, particularly for real-time racetrack analytics, has been instrumental in securing multi-million dollar deals. Currently, Nitin serves as the Data Governance Lead at MediaMarktSaturn, where he continues to drive data excellence and governance initiatives.
Vortrag Teilen
