கKatralAI

Campus Katral · Lesson 7 · 10 min

Neighbours in meaning-space

In one sentence

Embeddings put text on a map. Nearby points mean similar ideas — that is how search finds the right page.

Nila

Nila

Campus mentor

AI is not a person. It is a next-word engine with a very good memory of patterns.

Read the idea, watch the picture, then try it with your hands.

Watch the idea

The concept

Text

A citable chunk

Vector

A point on the meaning map

Neighbour

Similar idea, not keyword

A map of meaning

Embeddings locate text as vectors. Nearby points share ideas, not just keywords.

Chunk by idea

Random length cuts make bad neighbours. Split so a citation could stand on its own.

Worked example

Office hours vs hours in an office

Keyword search confuses them. A meaning map should keep them apart.

Picture to keep

Text → Vector → Neighbour

  1. 1Embed the chunk
  2. 2Measure distance
  3. 3Retrieve neighbours