Turn Unstructured Text into a Graph
80% of your organisation's data is unstructured, and hence unused.
Take unstructured, raw text data and easily build a knowledge graph that with all detected entities and relationships, and leverage more out of your text. In just a few clicks, with no code involved.
Discover how AP-HP turned patient data into a goldmine of information
How AP-HP uses knowledge graphs to structure patient data
See how it works in a quick video
See how it works in a 1-minute video 👇🏼
Build and enrich your knowledge graph.
With Lettria, you can find entities and relationships in unstructured data and automatically enrich your knowledge graph with more information.
While most graphs are built using structured, tabular data, with Lettria you can go one step ahead and start leveraging the rest of the data in your business.
Use Lettria's highly developed ontology, or use your own.
Use our internal ontology — over 1,000,000 words and concepts about the world — to build your knowledge graph.
Or you can use your own organisation's ontology for more custom use cases. We accept all standard formats.
Use the Graph DB that works for you.
While we use neo4j to visualise the graph on our platform, you can theoretically use any platform you like — TigerGraph, ArangoDB, MongoDB, Amazon Neptune — we can adapt to you.
We are platform-agnostic. The important thing is your access to your knowledge.
Your text made into a graph, step by step:
Here's what we offer in four steps:
1. Raw text import
The first step is to import all the your raw text that you want to use to build your graph.
2. Ontology import
The next step is to import your ontology — including the exact types of relationships and nodes that you want to identify.
3. Ontology alignment
We'll then align your ontology with the one we have internally, to make sure that we use the best of both for maximum accuracy.
4. Text to graph processing
Finally, we'll build the graph database based on all the information in your raw text.
See our text to graph demo in action
We've built a text to graph demo that's already live on our platform.
Here's what it looks like.
Input any raw text, get a graph out
With Lettria's text to graph demo, you can input any raw text and automatically generate a knowledge graph out of it.
Click on the image to see more.
Scan a business article, get deeper information
With Lettria's text to graph demo, you can gather deeper information and insights out of business articles.
Click on the image to see more.
Input a story, understand relationships
With Lettria's text to graph demo, you can input a story and get easy to understand information on characters, relationships, events, and objects.
Click on the image to see more.
Input a medical record, get more context
With Lettria's text to graph demo, you can input a medical record and access deeper and more detailed context about patient's history.
Click on the image to see more.
Read more about the Text to Graph Pipeline
Lettria Knowledge Studio