Eedi Blog

All blog posts

A unified knowledge graph

In London in 1854, Cholera was ravaging Soho. At the time, medicine believed that disease spread due to a “miasma” of pollution and bad air. The germ theory of disease had not yet been developed, and people were dying left and right.

A doctor named John Snow (no relation to Game of Thrones) was a skeptic of the prevailing theories and interventions. 

Our John Snow didn’t just focus on the symptoms. He mapped the cases, identifying a cluster around the broad street pump. Aided with new techniques of chemical and microscopic examination, John Snow did something radically different. He removed the pump’s handle. Unlike other doctors, he tackled the root cause of the contaminated water, shifting from treating the symptoms to solving the underlying cause.

Medicine since then has made incredible advances by understanding the cellular and genetic root causes of disease. But in education, we often find ourselves in a Pre-John Snow Era. We see symptoms - disengagement, underperformance, confusion, frustration.. 

At Eedi, we spent the last 10 years conducting basic research in the domain of Math. By decomposing math into fine grained constructs, and also understanding the misconceptions associated with specific constructs, we can identify the root cause. Rather than the broad interventions we apply, we can precisely diagnose that the part of a standard that’s holding a 9th grader back is tied to a misconception that the student picked up in 2nd grade. This work on a unified knowledge graph, led by organizations like CZI and ours, will help us remove the handle.

And there is still a tremendous amount of work to be done. I’m sure that many of you have heard that old adage that “math is math”. Well it turns out that given our experience in the UK and the US, it’s more like Math is often, but not always, “maths”. At Eedi, we are continuing this basic research to develop an open source, worldwide effort to develop these graphs across countries, languages, and context. With the support of Gates, Prevail, and Dell, we hope to bring the benefit of these graphs to 1 Bn kids by 2030, starting with India, Brazil, and Kenya. 

By understanding and using these graphs, we can stop treating the symptoms, and start addressing the root causes. We can help everyone be a little more like John Snow.

Written by

More from the Eedi Blog

Eedi’s Magic Number 🪄

Why 120 questions per year can transform maths progress in your class... At Eedi, teachers often ask: *“How many quizzes do my students really need to complete to meaningfully improve their maths?”* Thanks to a rigorous, externally conducted study, we now have a clear answer: students significantly benefit when they complete at least **120 Eedi check-in questions** over the course of a school year. Here’s what this means in practice — and why it matters for you and your students.

How Eedi drives meaningful maths gains—in just 10 minutes a week

Independent research confirms: Eedi delivers maths improvements equivalent to approximately 3–4 months of additional learning—with just 10 minutes of weekly engagement.

From Data to Discovery: Lessons Learned from Hosting AI Competitions in EdTech

At Eedi, we’re constantly pushing the boundaries of how data and AI can improve learning outcomes for students—and this time, we’re taking it to the next level. Simon Woodhead, Eedi’s Chief Data Scientist, shares an exciting behind-the-scenes look at how we’re using AI and competitions to tackle one of education’s toughest challenges: identifying and addressing student misconceptions in mathematics.