Data Wrangling/ Data Munging

Glossary page

Data munging, also known as data wrangling, refers to the manual process of converting or mapping data from its original "raw" form into a format that is more easily consumable with the aid of semi-automated tools. This process may involve additional steps such as data visualization, data aggregation, statistical modeling, or other applications. Typically, data munging follows a set of general steps, starting with extracting raw data from a source, using algorithms or parsing techniques to transform the data into predefined structures, and ultimately depositing the resulting content into a storage location for future use. With the rapid expansion of the internet, these techniques are becoming increasingly critical for organizing the vast amounts of available dat

https://online.hbs.edu/blog/post/data-wrangling external-link

Latest webinars

Latest articles

Blog Post Cover

Connect & Integrate: Simplifying certificate management with AI

Connect & Integrate brings the T-Systems AI power within the solution to support organizations scale up and become efficient in their business workflows within Catena-X. The cross-use-case AI layer sits at the intersection of data preparation and exchange as a business enabler directly within the workflows.

Read more

external-link
Author image

Tushar Yadav​

Aug 03, 2026

Blog Post Cover

IDSA’s Dataspaces and AI: Our input from applied Physial AI in RoX with dataspaces

The IDSA position paper "Data Spaces and AI" argues that the coordination problems facing agentic AI — identity, trust, access control, governance and provenance across organizational boundaries — are the same ones dataspaces already solve for cross-company data sharing. Rather than reinventing them for autonomous agents, the peer-to-peer logic behind standards like ISO/IEC 20151 can transfer directly to agent-to-agent interaction. T-Systems contributes a concrete proof point: RoX, a German consortium where competing robotics players and institutes such as DFKI and DLR share a sovereign data foundation for AI-based robotics. Built on Tractus-X and aligned with IDSA and Gaia-X, it already powers live robotic cells and is now extending that governed data supply to an agentic layer — showing in practice what the paper argues in principle.

Read more

external-link
Author image

Chris S. Langdon

Jul 21, 2026

Blog Post Cover

First Korea–Europe peer-to-peer dataspace transaction: L&F and EU Tier 1

In November 2025, the first peer-to-peer dataspace transaction between Korea and Europe was completed, with L&F (Korea) sharing product carbon footprint data with the automotive division of a major European Tier 1. Orchestrated by Prof. Chaisung Lim of Konkuk University and enabled by T-Systems' Dataspace-as-a-Service, the exchange ran on the Eclipse Tractus-X stack with the IDSA protocol and Gaia-X trust framework — the same standards underpinning Catena-X. The transaction proves that sovereign, governed data sharing can now travel across regions and regulatory environments without centralising data or surrendering control. Beyond secure, trusted file transfer, it points to a strategic foundation for AI-ready data and new value-creation scenarios, including controlled collaboration across ecosystems and even between competitors.

Read more

external-link
Author image

Chris S. Langdon

Jun 29, 2026