Data Governance - Knowledge Catalog

Today is an exciting milestone for IBM Knowledge Catalog with our latest software release, now generally available. Through 2023, our focus was on delivering value for our clients looking to IBM Knowledge Catalog who were most familiar with IBM InfoSphere Information Governance Catalog and Information Analyzer, as well as a set of data migration utilitizies for those clients. And we’re already seeing clients who made the decision successfully complete their move to IBM Knowledge Catalog. With the latest release, we have focused on three key themes:
1) Infusing generative AI into the product to drive productivity and improvements in metadata enrichment,
2) metadata visualization with an all-new relationship explorer designed to show the relationships between data assets and their governance artifacts, and
3) continued features for usability, including bulk actions on glossary and catalog assets via UI, analyzing assets with SQL-filtered data, column-level classifications and many more. Let’s take a look at some of these items in detail.

First is the release of new generative AI features, available with new IBM Knowledge Catalog Premium Cartridge, enhancing metadata enrichment. Many forward-thinking organizations are embracing self-service AI and Analytics to generate more business value from their data. But organizations can’t just give data teams unfettered access. The data could be uninterpretable, out of date, or too sensitive to use. At the end of the day, usable data needs to be accompanied by rich business and technical context that follows the data from storage, to processing, to cataloging, and, finally, to Analytics.  These new capabilities use IBM Research’s trusted slate 30m and granite 13b large language models (LLMs) to create a unified data context for IBM Knowledge Catalog. The new metadata enrichment with LLMs analyzes data sources to determine the meaning of each column of data. It deciphers this by using a variety of machine learning techniques to classify the actual data while analyzing column properties in relation to the context of the entire table.  The result is an enriched business context in the form of an expanded column name, column description, and business term assignment. This metadata context can then follow the data as it is transformed and consumed ensuring unified data understanding across tools and use cases.  Once assigned, this unified metadata context opens a whole world of automation and usability potential.  The new business context now enables more accurate searchability for the data and can be used to streamline access control and reporting downstream. To learn more, join us on July 10 for a webinar hosted by product manager, Corey Keyser, focused on this topic, “Harness the power of generative AI to scale data governance”

The other key feature I’d like to draw your attention to is the all-new relationship explorer, a tool that empowers data consumers and stewards to explore complex relationships between assets and governance artifacts, with the potential to dramatically improve impact analysis and data governance while enhancing data literacy across the organization. Inadequate data literacy can hinder an organization’s transformation to become more insights driven. A 2021 Forrester Trends report found that expanding companywide data literacy is a critical priority for 26% of global purchase influencers surveyed. And according to a 2023 Gartner report, poor data literacy in organizations perpetuates “gut feel” over data-driven decision making. Data consumers such as business analysts and data scientists often struggle to navigate the intricate web of relationships between assets and business metadata. A lack of visibility into these relationships can hinder data stewards’ ability to assess the impact of changing governance artifact assignments to data assets. This can impede governance and compliance efforts that require data stewards and compliance officers to identify the location of sensitive data, and manage governance policies and rules. Existing tools lack the capability to provide a comprehensive view of these relationships, leading to inefficiencies and compliance risks. Relationship explorer harnesses the power of a knowledge graph to provide a visual map that allows data consumers and stewards to effortlessly explore relationships between various data assets and governance artifacts. With fine-grained control over displayed content, comprehensive filters, and multiple entry points to initiate exploration from the appropriate context, relationship explorer serves as a unified interface for viewing relationships across all data assets and its metadata, increasing data literacy level in the organization. One sponsor user said, “When you’re crafting a business term and putting it together, at the end of the day, it’s not in isolation. It’s there for a reason. And I’m seeing “the reason”. I’m loving what you guys are showing.” Join us on June 20 for a webinar hosted by product manager, Michal Szylar, Supercharge data governance and data literacy with IBM’s Relationship Explorer

And there is much more. Check out the What’s new section of the documentation and contact your IBM or Partner Representative to learn more about IBM Knowledge Catalog.