curiosity

Whitepaper · Search scores

What a search score means

A plain language guide to the number next to each search result in Curiosity Studio: what it is, why it is not a percentage, and how to use it.

For Anyone who reads ranked results Chapters 6 Pages 10 Date October 2026

What it covers

A score puts results in order.

It does not say how good any one result is. A score is built from several parts, each with its own scale, so the number has no fixed top and cannot be read as a percentage. Use the order and the gaps, and open the result to judge it.

  1. Part 01

    Search terms

    Rare words count more

  2. Part 02

    Text similarity

    Closer meaning scores higher

  3. Part 03

    Graph links

    Shared connections

  4. Part 04

    Recency

    Last 14 days keep full score

  5. Part 05

    Boosts

    Chosen words or sources

Chapters

Six chapters, from a soup contest to three habits

01 · Page 02Chapter 01

A soup contest

What an order can and cannot tell you.

02 · Page 03Chapter 02

A sorting tool

A score puts results in order, and the bar is drawn against the best one.

03 · Page 04Chapter 03

How a score is made

Search terms, text similarity, graph links, recency and boosts.

04 · Page 05Chapter 04

Why it is not a percentage

Four reasons, each with an everyday example.

05 · Page 06Chapter 05

What a score can tell you

The questions a score can and cannot answer.

06 · Page 07Chapter 06

Reading a list

Three habits for reading ranked results.

The full paper

Read the order. Judge the result.

Download the PDF

Your data. Your infrastructure.

Connected knowledge for AI systems