The big idea: Decomposition, pattern recognition, abstraction and algorithmic design are not a programming method.
They are how any hard problem in computer science gets made solvable — and you have already used all four in this course.
You have met them already: Normalising a database was decomposition. An ERD was abstraction. A truth table was pattern recognition.
This micro names what you were doing, so you can do it deliberately.
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.
Database design
Decomposition splits one wide table into several. Abstraction decides an entity is only its relevant attributes. Pattern recognition notices that every many-to-many needs a linking table.
Normalising to 3NF is computational thinking with a different name.
Data analysis
Abstraction decides which columns of a huge dataset the question actually needs. Pattern recognition finds the trend or the grouping.
Decomposition splits "why did sales fall?" into questions that each have an answer.
Machine learning
Abstraction chooses which features to give the model and which to leave out — the single most important decision in a project.
Pattern recognition is literally what the model does; the person's job is deciding what patterns would be meaningful.
Network security
Decomposition splits a network into segments so a problem is contained. Pattern recognition spots traffic that does not look like normal use.
Algorithmic design writes the firewall rules, in an order that produces the intended result.
Why this matters for the exam: Paper 2 sets a question on algorithmic thinking with no code to read or write.
It is testing exactly this: can you take a problem in words and make it solvable, without reaching for a programming language.
Get feedback like a real examiner
Submit your answers and get instant feedback — what you did well, what's missing, and exactly what to write to score full marks.
A method for unfamiliar problems: When a question describes something unfamiliar, the four concepts are a procedure, not a topic.
Apply them in order and an unfamiliar problem becomes several familiar ones.
1 — What are the pieces?
Write down the separate jobs the system must do. Do not solve any of them yet.
If a piece still feels too big to describe in a sentence, break it again.
2 — What repeats?
Look across the pieces for anything with the same shape.
Anything you would write twice is a sign that one solution can cover both — and that is usually where a loop or a function belongs.
3 — What can I ignore?
For each piece, list what it genuinely needs to know, and be ruthless about the rest.
Detail you keep "just in case" is detail you will have to handle later for no benefit.
4 — What are the steps?
Write the steps in order, in plain sentences. Then trace them on a small example by hand.
If the trace gives the wrong answer, the algorithm is wrong — and you have found that out before writing a line of code.
The trace is not optional: Working a small example by hand is what turns a plausible-looking algorithm into one you know is right.
It is also, very often, exactly what the exam asks you to do.
How this is tested — you must apply the concepts to an unfamiliar problem and show the thinking, not write code. It comes up two ways:
Paper 2 — working with code
- Explain how the concepts solve a described problem, 4-6 marks
- Break an unfamiliar problem into parts
- Justify what you chose to ignore
Paper 2 — the algorithmic-thinking question
- The algorithmic-thinking question is set here deliberately
- No code to read or write — the reasoning is the answer
The classic trap: Jumping to a solution. The marks are for the route — the pieces you identified, the pattern you spotted, what you left out and why. A right answer with no reasoning scores poorly here.
A city wants to know which of its 400 bus stops should get a shelter first. Explain how computational thinking would be applied. No code is required.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.