Personalised Learning — Flagship Feature

Every child's next lesson is the right one.

ClassGrade doesn't guess at personalisation. It uses information theory to compute, with mathematical precision, exactly what each student should learn next — and why.

Knowledge Graph

A living map of every student's understanding

Each node is a concept. Each edge is a prerequisite. Colour intensity shows mastery. Pulsing nodes are high-entropy — the system's next priority.

Mastered (≥80%)
In Progress (40–79%)
Needs Attention (<40%)
Pulsing = high entropy (next lesson candidates)
95%Phonics91%Sight Words74%Vocabulary83%Grammar89%Spelling78%Reading Fluency60%Comprehension43%Writing Mechanics51%Sentence Structure31%Inference24%Essay Writing14%Literary Analysis★ NEXT RECOMMENDED

Hover nodes to inspect mastery & entropy · Pulsing = high information-gain opportunities

Example: This graph shows one student's English comprehension map. Foundation concepts (bottom) are mostly mastered. Advanced concepts (top) are high-entropy — the system will focus here next.

300K+Students on the graph
<10msNext-lesson decision
Neo4jGraph database
Unique learning paths
The Learning Loop

Observe. Analyse. Decide. Generate. Learn.

Every session closes the loop. Every response sharpens the model. Every student gets a path that is uniquely theirs.

1

Observe

Every student interaction is captured and linked to the relevant concept node in the knowledge graph.

2

Analyse

The system recomputes each concept's mastery estimate and information-theoretic entropy after every submission.

3

Decide

The concept with the highest entropy — weighted by prerequisite readiness — is selected as the next focus.

4

Generate

An activity targeting that concept is retrieved from the library or generated on demand.

5

Learn

The student engages with the activity. The response updates the mastery estimate. The loop restarts.

ClassGradeAI EngineObserveEvery interaction capturedAnalyseEntropy computed perDecideOptimal next activityGenerateContent created orLearnStudent engages &
Why Entropy

Traditional sequencing vs. entropy-based routing

The difference is not cosmetic. It's the difference between a fixed path and a model that knows what it doesn't know.

Traditional Sequencing

  • All students follow the same content order
  • Advancement based on completion, not mastery
  • Fast learners are held back; struggling students fall further behind
  • No mechanism to distinguish "hasn't seen it yet" from "doesn't understand it"
  • Teacher must identify gaps manually

Entropy-Based Routing

  • Each student follows a unique path based on their knowledge state
  • Advancement only when mastery estimate crosses a confidence threshold
  • High-entropy concepts (maximum uncertainty) get immediate attention
  • Prerequisite graph ensures no concept is targeted before its foundations are solid
  • Gaps surface automatically — no teacher analysis needed

See the knowledge graph for your curriculum

We'll build a live knowledge graph from your content and show you the personalisation engine in action.