The big idea: OLAP answers questions you already have. Data mining finds patterns you did not know to look for.
The difference is who supplies the hypothesis — you, or the data.
| OLAP | Data mining | |
|---|---|---|
| You start with | A question | No question — just the data |
| It gives you | The answer to that question | Patterns you did not ask about |
| Driven by | A person exploring | An algorithm searching |
| Example | "Sales by region by quarter?" | "These two products sell together" |
| Result is | A fact | A candidate finding, needing checking |
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
The cube: OLAP data is modelled as a cube: measures (sales, units) across dimensions (time, product, region, store).
Every OLAP operation is a way of moving through that cube.
Drill down
Move to finer detail — year to quarter to month, or region to store.
Roll up
The reverse: aggregate to a coarser level. Store to region to country.
Slice
Fix one dimension to a single value — "only 2025" — leaving a smaller cube.
Dice
Fix several dimensions to ranges — "2024–25, in Europe, for drinks".
Pivot
Rotate the view so a different dimension forms the rows or columns. The same numbers, read a different way.
Drill down and roll up are opposites: Down to detail, up to summary. Slice fixes one dimension, dice fixes several.
These four are the most commonly asked definitions in the topic, and they are routinely swapped.
Learn what examiners really want
See exactly what to write to score full marks. Our AI shows you model answers and the key phrases examiners look for.
What mining looks for
- Association — things that occur together
- Clustering — groups nobody defined
- Classification — rules predicting a known label
- Anomaly detection — records unlike the rest, which is how fraud is found
- Regression — predicting a number from other values, such as next month's sales
- Sequential pattern discovery — what tends to follow what, over time
Search enough and you will find something: Test thousands of combinations and some will look significant by chance alone. A found pattern is a candidate, not a finding — it must be checked on data that was not used to find it.
This is the central honesty problem of data mining, and it is examinable.
And it is never causation: Mining reports that two things co-occur. Acting as though one causes the other is a decision the data does not support — and is where most real-world data-mining harm comes from.
How this is tested — you must distinguish answering a question from discovering a pattern, and name the OLAP operations exactly. It comes up two ways:
Paper 1 Section A
- Explain the role of OLAP and data mining, 3-4 marks
- Define drill down, roll up, slice or dice
- State one data mining technique
Paper 1 Section B — case study
- Choose between OLAP and mining for a task
- Explain a limitation of a mined pattern
The classic trap: Treating a mined pattern as a fact. It is a candidate — search enough combinations and chance supplies some — and it reports co-occurrence, never cause.
The cube, and the operations an analyst uses on it.
Interactive diagram
Explore the labelled diagram, charts and maps for this topic in full study mode.
A retailer has a warehouse of five years of sales. Explain how OLAP and data mining would each be used, and state one limitation of the mining results.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.