Connecting the Dots: How Knowledge Graphs Enhance Content

Sign up for our newsletter on globalization and localization matters.

Teodora Petkova (opens to a new tab)  is Knowledge Graph Content and Data Manager at Ontotext (opens to a new tab). She a philologist fascinated by the metamorphoses of text on the Web and curious about the ways the Semantic Web unfolds. Teodora holds a PhD. in Marketing Communication, an MS in Creative writing and a Bachelor of Science in Classics. With 15+ years of experience working as a content writer, she fully and wholeheartedly focuses on digital marketing communication that create dialogic moments through semantic annotations.

In our evolving digital landscape, the power of knowledge graphs continues to revolutionize industries. For the localization industry, understanding the concept of knowledge graphs can significantly enhance your approach to managing multilingual content and improving decision-making processes. 

Watch the recorded interview above and read the recap underneath for a primer on knowledge graphs and their applications.

What Is a Knowledge Graph?

A knowledge graph is an organized structure that connects disparate pieces of information, much like a well-organized library connects related books, a curiosity cabinet of sorts. Instead of relying on traditional databases that categorize data in isolation, knowledge graphs allow entities such as products, terms, and cultural references to be meaningfully connected. This connectivity allows you to discover, retrieve, and understand information more effectively.

Using Semantic Web technologies, knowledge graphs bridge the gap between human understanding and machine-readable data, making them essential tools for improving business processes in the localization industry.

Why Knowledge Graphs Matter in Localization

Knowledge graphs are becoming increasingly important for localization by driving improvements in several key areas:

1. Improving multilingual content management: Knowledge graphs excel at connecting terms and concepts across languages. Multilingual content can be updated consistently, ensuring accurate and consistent translations that resonate with different audiences.

2. Improve knowledge management: Effective terminology management has a direct impact on the success of translation projects. Knowledge graphs enable localization teams to manage terminology consistently, linking different synonyms and contextually relevant phrases to maintain brand voice and integrity across markets.

3. Inform business decisions: With knowledge captured and connected, organizations can make data-driven decisions based on robust analytics. By leveraging knowledge graphs, teams can derive insights and gain clarity about their offerings and audiences.

4. Improve content discovery and accessibility: Imagine a workplace where you can seamlessly search for relevant information across a myriad of documents and terms. Knowledge graphs facilitate efficient searches tied to specific queries, allowing localization professionals to quickly access the insights they need.

Steps to Building a Knowledge Graph for Localization

Starting a knowledge graph for your localization efforts may seem daunting, but breaking it down into manageable steps can simplify the process:

1. Define goals and purpose: Clearly define what you want your knowledge graph to accomplish. Whether it's facilitating translation or improving content retrieval, identifying goals is key.

2. Formulate competency questions: Generate questions that will guide the design of your knowledge graph. As localization professionals, questions might include aspects such as: "What concepts are most often associated with product translations?"

3. Leverage existing data: Gather relevant information from your current systems (e.g., content databases, glossaries). Consider how keywords relate to specific markets and consumer behavior.

4. Establish relationships: Clearly map the connections and associations between your terms. Define synonyms, translations, and culturally relevant contexts. This enriched data helps build a robust localization engine.

5. Iterate and refine: Building a knowledge graph is an iterative process. Based on user feedback and performance, you can adjust and improve the structure of your knowledge graph over time.

Resources Mentioned in the Interview

T. Petkova, Metaphors To Think Knowledge Graphs By (opens to a new tab)

T. Petkova, If Curiosity Cabinets Were Knowledge Graphs (opens to a new tab)

V. Alexiev, 10 Steps Of Building A knowledge Graph (opens to a new tab)

Database of Known Fakes (opens to a new tab)

Ontotext's Demonstrator (opens to a new tab)

Taming Content Complexity with a Knowledge Graph (opens to a new tab)

Maria Keet, The What and How of Modelling Information and Knowledge (opens to a new tab) 

Do you want to contribute with an article, a blog post or a webinar?

We’re always on the lookout for informative, useful and well-researched content relative to our industry.

Write to us.