What each instruction word expects in your answer — the single biggest reason marks are lost is answering a "describe" question with a "state"-level response.
Define — give the precise meaning of a word or concept.
State — give a specific name, value or brief answer — no explanation needed.
Identify — pick out an answer from a number of possibilities.
Outline — give a brief, structured summary or overview.
Describe — give a detailed, step-by-step account.
Explain — give a detailed account that includes reasons or causes.
Distinguish — make the differences between two or more items clear.
Compare — give an account of the similarities between two or more items.
Construct — display information in a logical/diagrammatic form, or build the code/query/diagram asked for.
Discuss — give a balanced review that includes a range of arguments or factors.
Evaluate — weigh up strengths and limitations to reach a conclusion.
Suggest — propose a possible answer, solution or hypothesis.
B1 · Computational Thinking
Abstraction — hiding unnecessary detail to focus on what matters for the problem.
Decomposition — breaking a complex problem into smaller, manageable parts.
Pattern recognition — spotting similarities or trends across a problem to reuse a known solution.
Success criteria — specific, testable statements that define when a solution counts as working.
B2 · Programming
Iteration — repeating a block of code, using a counted or conditional loop.
Selection — choosing between code paths based on a condition (if/elif/else).
Stack — a LIFO (last-in, first-out) structure — only push/pop the top item.
Queue — a FIFO (first-in, first-out) structure — items leave in the order they arrived.
Big-O notation — a way of describing how an algorithm's run time grows as input size grows.
Exception — a run-time error that can be anticipated and handled with try/except.
B3 · Object-Oriented Programming
Class — a blueprint defining the attributes and methods an object will have.
Object — a specific instance created from a class.
Encapsulation — bundling data with the methods that operate on it, restricting direct outside access.
Inheritance — a subclass reusing and extending the attributes/methods of a parent class.
Polymorphism — the same method name behaving differently depending on the object calling it.
Composition/aggregation — modelling "has-a" relationships between objects, with or without shared lifetime.
B4 · Abstract Data Types (HL)
Abstract data type (ADT) — a data structure defined by its operations, not its internal implementation.
Linked list — a sequence of nodes, each pointing to the next, allowing flexible insertion/removal.
Binary search tree — a tree where each node's left subtree is smaller and right subtree is larger.
Hashmap — a structure that maps keys to values using a hash function for near-instant lookup.
A1 · Computer Fundamentals
Fetch–decode–execute cycle — the repeating process a CPU uses to run each instruction in memory.
Register — a small, very fast storage location inside the CPU.
Pipelining — overlapping stages of multiple instructions to increase throughput.
Logic gate — a circuit implementing a basic Boolean operation (AND, OR, NOT, etc).
Scheduling — the OS deciding which process gets the CPU, and for how long.
Interrupt — a signal that pauses the CPU to immediately handle an urgent event.
A2 · Networks
Topology — the physical/logical layout of how devices connect on a network.
Protocol — an agreed set of rules that lets devices communicate correctly.
Packet switching — splitting data into packets sent independently and reassembled at the destination.
Firewall — a system that filters network traffic against a set of security rules.
Encryption — scrambling data so only someone with the correct key can read it.
A3 · Databases
Primary/foreign key — a primary key uniquely identifies a row; a foreign key links to another table's primary key.
Entity-relationship diagram — a diagram modelling entities and how they relate, before building tables.
Normalization — organizing tables to reduce redundancy and avoid update anomalies.
SQL join — combining rows from two or more tables based on a related column.
Data warehouse — a central store consolidating data from multiple sources for analysis.
A4 · Machine Learning (HL)
Feature — an individual measurable input variable used by a model.
Regression — predicting a continuous numeric value from input data.
Classification — assigning input data to one of a set of predefined categories.
Overfitting — a model learning the training data too closely, hurting performance on new data.
Neural network — layers of connected nodes that transform input into an output through weighted connections.
Case Study · Generative AI for Image Creation
Diffusion model — a model that generates images by learning to reverse a noise-adding process.
GAN — a generator and discriminator trained against each other to produce realistic output.
Hybrid model — an approach combining diffusion and GAN techniques to balance quality and speed.
Want the full course behind these terms?
Lesson decks, revision handouts, exam-style practice with worked answer keys, and AI-powered exam prep — one payment, one year of access.
See what's included →