📚 A-Level AQA Computer Science: Mind Map Quick Revision | A-Level AQA 计算机：思维导图速记
A mind map is a powerful study tool that organises key concepts visually, helping you link ideas and remember the AQA A-Level Computer Science syllabus efficiently. This article presents a structured mind map overview covering all major topics, with each branch broken down into bite-sized points for rapid revision.
思维导图是一种强大的学习工具，可以直观地组织关键概念，帮助你高效地串联和记忆 AQA A-Level 计算机科学课程内容。本文提供一个结构化的思维导图概览，覆盖所有主要话题，每个分支都细化为便于快速复习的知识点。
1. Programming Fundamentals | 编程基础
Programming fundamentals form the core of writing effective code, covering data types, variables, and constants that store information during execution.
编程基础是编写有效代码的核心，包括数据类型、变量以及在程序执行时存储信息的常量。
Control structures direct the flow of execution: sequence, selection (if-else, switch/case), and iteration (for, while, do-while loops) are universal building blocks.
控制结构指引执行流程：顺序、选择（if-else, switch/case）和迭代（for、while、do-while 循环）是通用的构建模块。
Subroutines (procedures and functions) promote reusability and modular design; parameters can be passed by value or by reference, and functions return a single value using a return type.
子程序（过程和函数）促进代码重用和模块化设计；参数可以按值或按引用传递，函数通过返回类型返回一个值。
Exception handling with try-catch blocks prevents runtime crashes by intercepting errors gracefully and allowing corrective action.
使用 try-catch 代码块的异常处理可以优雅地拦截错误并允许采取纠正措施，防止运行时崩溃。
Recursion is a technique where a subroutine calls itself to solve a problem by reducing it to a smaller instance, requiring a base case to avoid infinite loops.
递归是一种子程序调用自身来解决问题的技术，将问题简化为更小的实例，必须设置基准情形以避免无限循环。
Object-oriented programming (OOP) introduces classes, objects, inheritance, encapsulation, and polymorphism to model real-world entities and relationships.
面向对象编程（OOP）引入了类、对象、继承、封装和多态，用来模拟现实世界的实体和关系。
File handling allows programs to read from and write to external files; common operations include open, close, read, write, and append, often using text or binary modes.
文件处理允许程序读写外部文件；常见操作包括打开、关闭、读、写和追加，通常使用文本或二进制模式。
2. Data Structures | 数据结构
Arrays are fixed-size, indexed collections of elements of the same data type, enabling constant-time O(1) random access but inefficient insertion/deletion.
数组是固定大小、带索引的同类型元素集合，支持常数时间 O(1) 的随机访问，但插入和删除操作效率较低。
Lists (dynamic arrays) can grow and shrink, offering flexibility at the cost of occasional resizing overhead; they are fundamental in Python as ‘list’ and in Java as ‘ArrayList’.
列表（动态数组）可以动态扩容和缩容，提供了灵活性，但偶尔会产生重新分配的开销；在 Python 中表示为 list，在 Java 中为 ArrayList。
Stacks follow Last-In-First-Out (LIFO), supporting push and pop operations; used in function call management, undo mechanisms, and expression evaluation.
栈遵循后进先出（LIFO）原则，支持 push 和 pop 操作；用于函数调用管理、撤销操作和表达式求值。
Queues are First-In-First-Out (FIFO) with enqueue and dequeue operations; priority queues and circular queues adapt the idea for scheduling and buffering.
队列是先进先出（FIFO）结构，具有入队和出队操作；优先队列和循环队列将这种思想用于任务调度和缓冲。
Linked lists consist of nodes containing data and a pointer to the next node; singly, doubly, and circular variants provide dynamic memory usage and efficient insertion/deletion.
链表由包含数据和指向下一个节点的指针的节点组成；单向、双向和循环变体提供了动态内存使用和高效的插入/删除。
Trees are hierarchical structures with a root node and child nodes; binary search trees (BSTs) allow O(log n) search, insertion, and deletion when balanced.
树是具有根节点和子节点的层次结构；平衡的二叉搜索树（BST）可实现 O(log n) 的搜索、插入和删除操作。
Hash tables map keys to values via a hash function, offering average O(1) lookup; collisions are resolved through chaining or open addressing.
哈希表通过哈希函数将键映射到值，提供平均 O(1) 的查找速度；冲突通过链地址法或开放地址法解决。
Graphs model networks with vertices and edges; can be directed/undirected and weighted/unweighted; represented via adjacency matrices or adjacency lists.
图用顶点和边来建模网络；可以是有向/无向和加权/无权的；通过邻接矩阵或邻接表表示。
3. Algorithms | 算法
Searching algorithms include linear search (O(n)) and binary search (O(log n) on sorted data); binary search repeatedly divides the search interval in half.
搜索算法包括线性搜索（O(n)）和二分搜索（在有序数据上的 O(log n)）；二分搜索不断将搜索区间对半分。
Sorting algorithms: bubble sort (O(n²)) repeatedly swaps adjacent out-of-order elements; merge sort (O(n log n)) uses divide-and-conquer by splitting and merging sorted halves.
排序算法：冒泡排序（O(n²)）反复交换相邻逆序元素；归并排序（O(n log n)）采用分治法，将数组分成两半排序后再合并。
Graph traversal: depth-first search (DFS) uses a stack (or recursion) to explore as far as possible along each branch; breadth-first search (BFS) uses a queue to explore level by level.
图遍历：深度优先搜索（DFS）使用栈（或递归）沿着每个分支尽可能深入探索；广度优先搜索（BFS）使用队列逐层探索。
Shortest path algorithms: Dijkstra’s algorithm finds the shortest path from a single source in weighted graphs with no negative weights; A* uses heuristics to improve efficiency.
最短路径算法：Dijkstra 算法在没有负权重的加权图中找到单源最短路径；A* 使用启发式函数来提高效率。
Complexity analysis uses Big O notation to describe upper bounds of time or space requirements; common classes: O(1), O(log n), O(n), O(n log n), O(n²), O(2ⁿ).
复杂度分析使用大 O 符号描述时间或空间需求的上界；常见类别：O(1)、O(log n)、O(n)、O(n log n)、O(n²)、O(2ⁿ)。
Tractable problems can be solved in polynomial time; intractable problems may only have exponential solutions; the class P vs NP explores whether every problem whose solution can be verified quickly can also be solved quickly.
可解问题可以在多项式时间内解决；难解问题可能只有指数时间的解法；P vs NP 问题探讨是否每个可快速验证解的问题也能被快速求解。
Halting problem is undecidable: no general algorithm can determine whether any given program will finish running or loop forever.
停机问题是不可判定的：没有一个通用算法可以判断任意给定程序是会停止运行还是永远循环。
4. Theory of Computation | 计算理论
Finite state machines (FSMs) consist of states, inputs, outputs, and transitions; they model sequential logic and are used in lexical analysis and control systems.
有限状态机（FSM）由状态、输入、输出和转换组成；它用于建模时序逻辑，并应用于词法分析和控制系统。
Mealy and Moore machines are two types of FSMs: Mealy outputs depend on current state and input, Moore outputs depend only on state.
Mealy 机和 Moore 机是两种 FSM 类型：Mealy 型的输出取决于当前状态和输入，Moore 型的输出仅取决于状态。
Regular expressions define patterns for string matching, using operators like union (|), concatenation, and Kleene star (*) to describe regular languages.
正则表达式定义了字符串匹配的模式，使用并（|）、连接和克林星号（*）等运算符来描述正则语言。
Context-free grammars (CFGs) use production rules to generate languages, described in Backus-Naur Form (BNF); they are more expressive than regular languages, used to define programming language syntax.
上下文无关文法（CFG）使用产生式规则生成语言，以巴科斯-诺尔范式（BNF）描述；它比正则语言更具表达能力，用于定义编程语言的语法。
Turing machines provide a formal model of computation; they manipulate symbols on an infinite tape according to a set of rules, and can simulate any algorithm.
图灵机提供了计算的正式模型；它根据一组规则在无限长的纸带上操作符号，能够模拟任何算法。
The Church-Turing thesis states that any effectively calculable function can be computed by a Turing machine, establishing the boundaries of what is algorithmically possible.
邱奇-图灵论题指出，任何有效可计算的函数都可以由图灵机计算，这确立了算法可能性的界限。
5. Data Representation | 数据表示
Number systems: binary (base-2), hexadecimal (base-16), and decimal (base-10) conversions are essential; hexadecimal compactly represents binary nibbles.
数制：二进制（基数为2）、十六进制（基数为16）和十进制（基数为10）的转换是基础；十六进制可以紧凑地表示二进制半字节。
Unsigned and signed integers use two’s complement for negative numbers, where the most significant bit is the sign bit; range for n bits: –2ⁿ⁻¹ to 2ⁿ⁻¹–1.
无符号和有符号整数使用二进制补码表示负数，其中最高位是符号位；n 位的范围：–2ⁿ⁻¹ 到 2ⁿ⁻¹–1。
Floating-point representation uses mantissa and exponent (e.g., IEEE 754) to store real numbers; precision trades off with range; normalisation ensures unique representation.
浮点表示使用尾数和指数（如 IEEE 754）来存储实数；精度与范围需要权衡；规格化可确保表示的唯一性。
Character encoding: ASCII uses 7 or 8 bits for common characters; Unicode (UTF-8, UTF-16) supports global scripts and emojis, backwards compatible with ASCII.
字符编码：ASCII 使用 7 或 8 位表示常用字符；Unicode（UTF-8、UTF-16）支持全球文字和表情符号，并向后兼容 ASCII。
Images are stored as bitmaps (pixel grids) or vectors (geometric primitives); colour depth and resolution affect file size and quality; metadata stores dimensions and colour info.
图像以位图（像素网格）或矢量图（几何图元）形式存储；颜色深度和分辨率影响文件大小和质量；元数据存储尺寸和颜色信息。
Sound is sampled at a rate (Hz) and bit depth; Nyquist theorem states sampling rate must be at least twice the highest frequency to avoid aliasing; MIDI stores music as instructions.
声音以采样率（Hz）和位深度进行采样；奈奎斯特定理指出采样率必须至少为最高频率的两倍以避免混叠；MIDI 以指令形式存储音乐。
Data compression: lossless (run-length encoding, Huffman coding) retains all original data; lossy (JPEG, MP3) removes less perceptible information to drastically reduce size.
数据压缩：无损（游程编码、霍夫曼编码）保留所有原始数据；有损（JPEG、MP3）去除不易感知的信息，大幅度缩减大小。
6. Computer Systems & Organisation | 计算机系统与组成
The processor (CPU) executes instructions using the fetch-decode-execute cycle; it contains the control unit, ALU, registers (PC, MAR, MDR, CIR, ACC), and clock.
处理器（CPU）通过取指-解码-执行周期来执行指令；它包含控制单元、算术逻辑单元、寄存器（PC、MAR、MDR、CIR、ACC）和时钟。
Factors affecting CPU performance: clock speed (cycles per second), number of cores (parallel execution), and cache size/level (L1, L2, L3 reduce memory access latency).
影响 CPU 性能的因素：时钟速度（每秒周期数）、核心数（并行执行）和缓存大小/级别（L1、L2、L3 减少内存访问延迟）。
Memory hierarchy: registers (fastest, smallest) → cache → main memory (RAM) → secondary storage (HDD, SSD). RAM is volatile, ROM is non-volatile and stores firmware.
存储层次：寄存器（最快、最小）→ 缓存 → 主存储器（RAM）→ 辅助存储器（HDD、SSD）。RAM 是易失性的，ROM 是非易失性的，存储固件。
Secondary storage devices: magnetic (HDD) uses spinning platters; optical (CD/DVD) uses lasers; solid state (SSD) uses floating-gate transistors, offering faster access and no moving parts.
辅助存储设备：磁性存储（HDD）使用旋转盘片；光学存储（CD/DVD）使用激光；固态存储（SSD）使用浮栅晶体管，提供更快的访问速度且无活动部件。
The stored program concept (Von Neumann architecture) places both instructions and data in the same memory; Harvard architecture uses separate memories for instruction and data, enabling simultaneous access.
存储程序概念（冯·诺依曼架构）将指令和数据放在同一内存中；哈佛架构则使用独立存储体存放指令和数据，允许同时访问。
Operating systems manage hardware, provide a user interface, handle multitasking, memory management, peripheral management, and security (user accounts, file permissions).
操作系统管理硬件、提供用户界面、处理多任务、内存管理、外设管理和安全性（用户帐户、文件权限）。
BIOS performs POST and loads the bootloader; interrupts and system calls allow software to request OS services; direct memory access (DMA) enables devices to bypass CPU for data transfer.
BIOS 执行开机自检并加载引导程序；中断和系统调用允许软件请求操作系统服务；直接内存访问（DMA）使设备可以绕过 CPU 直接传输数据。
7. Communication & Networking | 通信与网络
Networking types: LAN (local area), WAN (wide area), PAN (personal area), and WLAN (wireless LAN); topologies include star, bus, ring, and mesh, each with trade-offs in cost and fault tolerance.
网络类型：LAN（局域网）、WAN（广域网）、PAN（个人区域网）和 WLAN（无线局域网）；拓扑结构包括星型、总线型、环形和网状，各在成本和容错方面有所权衡。
The TCP/IP stack model: application (HTTP, FTP, SMTP), transport (TCP, UDP), internet (IP), and link layers (Ethernet, Wi-Fi); encapsulation adds headers as data passes down layers.
TCP/IP 协议栈模型：应用层（HTTP、FTP、SMTP）、传输层（TCP、UDP）、网际层（IP）和链路层（以太网、Wi-Fi）；数据在下层被封装时添加头部。
IP addressing (IPv4: 32-bit, IPv6: 128-bit) uniquely identifies devices; subnetting and CIDR notation manage address space; DNS translates domain names to IP addresses.
IP 地址（IPv4：32位，IPv6：128位）唯一标识设备；子网划分和 CIDR 表示法管理地址空间；DNS 将域名转换为 IP 地址。
Packet switching breaks data into packets that may take different routes; routers forward packets based on routing tables; protocols like TCP ensure reliable delivery with acknowledgements and retransmission.
分组交换将数据分解为可能经过不同路由的数据包；路由器根据路由表转发数据包；TCP 等协议通过确认和重传确保可靠交付。
Network security: firewalls filter traffic; encryption (symmetric and asymmetric) scrambles data; digital signatures and certificates verify authenticity and integrity; SSL/TLS secures web communication.
网络安全：防火墙过滤流量；加密（对称和非对称）对数据加扰；数字签名和证书验证真实性和完整性；SSL/TLS 保护网络通信。
The client-server model has centralised servers responding to client requests; peer-to-peer models distribute responsibility among participants, improving resilience and scalability.
客户端-服务器模型采用集中式服务器响应客户端请求；对等网络（P2P）模型将责任分配给各个参与者，提高了弹性和可扩展性。
Web technologies: HTTP/HTTPS methods (GET, POST, PUT, DELETE) and status codes (200 OK, 404 Not Found); RESTful APIs use stateless request-response cycles and standard URLs.
网络技术：HTTP/HTTPS 方法（GET、POST、PUT、DELETE）和状态码（200 OK、404 Not Found）；RESTful API 使用无状态的请求-响应周期和标准 URL。
8. Databases & Big Data | 数据库与大数据
Relational databases store data in tables (relations) linked by keys; each row is a tuple, columns are attributes, and primary keys uniquely identify records; foreign keys establish relationships.
关系数据库将数据存储在由键关联的表中；每一行是一个元组，列是属性，主键唯一标识记录；外键建立关系。
Normalisation reduces redundancy by organising data into multiple related tables; 1NF enforces atomicity, 2NF removes partial dependencies, 3NF removes transitive dependencies.
规范化通过将数据组织到多个相关表中来减少冗余；1NF 强制原子性，2NF 消除部分函数依赖，3NF 消除传递函数依赖。
SQL (Structured Query Language) is used to define, manipulate, and query data; DDL (CREATE, ALTER, DROP), DML (SELECT, INSERT, UPDATE, DELETE), and DCL (GRANT, REVOKE) commands are essential.
SQL（结构化查询语言）用于定义、操作和查询数据；DDL（CREATE、ALTER、DROP）、DML（SELECT、INSERT、UPDATE、DELETE）和 DCL（GRANT、REVOKE）命令是基础。
Entity-relationship (E-R) modelling uses entities, attributes, and relationships (one-to-one, one-to-many, many-to-many) to design databases before implementation.
实体-关系（E-R）建模使用实体、属性和关系（一对一、一对多、多对多）在实现之前设计数据库。
Big Data is characterised by the three Vs: Volume (huge amounts), Velocity (high speed of generation), and Variety (structured, unstructured, semi-structured); additional Vs include Veracity and Value.
大数据的特征为三个 V：Volume（海量数据）、Velocity（高速生成）和 Variety（结构化、非结构化、半结构化）；还包括 Veracity（真实性）和 Value（价值）。
Distributed computing frameworks like MapReduce split large tasks across clusters; Hadoop and Spark enable processing of petabytes of data with fault tolerance through replication.
MapReduce 等分布式计算框架将大型任务拆分到集群中执行；Hadoop 和 Spark 通过复制实现容错，可处理 PB 级数据。
Data analytics uses machine learning algorithms to extract patterns; unsupervised learning (clustering) and supervised learning (classification/regression) are commonly applied to big datasets.
数据分析使用机器学习算法提取模式；无监督学习（聚类）和监督学习（分类/回归）常应用于大数据集。
9. Functional Programming | 函数式编程
Functional programming treats computation as the evaluation of mathematical functions, avoiding mutable state and side effects; pure functions always produce the same output for the same input.
函数式编程将计算视为数学函数的求值，避免可变状态和副作用；纯函数对于相同输入总是产生相同输出。
First-class functions allow functions to be assigned to variables, passed as arguments, and returned from other functions, enabling higher-order functions.
一等函数允许将函数赋值给变量、作为参数传递以及从其他函数返回，从而支持高阶函数。
Anonymous functions (lambda expressions) are function definitions without a name, expressed inline: often used with map, filter, and reduce/fold operations on lists.
匿名函数（Lambda 表达式）是没有名称的函数定义，以内联方式表达；通常与列表的 map、filter 和 reduce/fold 操作一起使用。
Immutability means data structures cannot be modified after creation; new versions are created with the desired changes, simplifying reasoning about program state and concurrency.
不可变性意味着数据结构创建后不可修改；需要变更时会创建新版本，这简化了程序状态和并发性的推理。
Recursion replaces iterative loops; a function calls itself with a smaller subproblem until a base case is reached; tail recursion can be optimised by the compiler to avoid stack overflow.
递归替代了迭代循环；函数使用更小的子问题调用自身直到达到基准情形；尾递归可以通过编译器优化避免栈溢出。
List comprehensions provide a concise way to create new lists by applying expressions to elements of existing lists, often combined with guards (conditions).
列表推导式提供了一种通过对现有列表元素应用表达式来创建新列表的简洁方式，常与条件守卫结合使用。
Common functional languages include Haskell, Lisp, and F#; AQA focuses on applying functional concepts within a multi-paradigm language such as Python or C#.
常见的函数式语言包括 Haskell、Lisp 和 F#；AQA 侧重于在多范式语言（如 Python 或 C#）中应用函数式概念。
10. Consequences of Uses of Computing | 计算应用的影响
Privacy concerns arise from data collection by apps, websites, and IoT devices; data protection laws (like GDPR) require consent, transparency, and the right to be forgotten.
应用、网站和物联网设备收集数据引发隐私问题；数据保护法（如 GDPR）要求获得同意、保持透明度并赋予用户被遗忘权。
Cybersecurity challenges include malware (viruses, worms, trojans, ransomware), phishing, and denial-of-service attacks; robust incident response and patching are essential.
网络安全挑战包括恶意软件（病毒、蠕虫、木马、勒索软件）、网络钓鱼和拒绝服务攻击；稳健的应急响应和补丁管理至关重要。
Artificial intelligence and automation impact employment, replacing repetitive tasks but also creating new jobs; bias in training data can lead to unfair algorithmic decisions.
人工智能和自动化对就业产生影响，替代重复性任务但也创造新的工作岗位；训练数据中的偏见可能导致不公平的算法决策。
Intellectual property issues involve software piracy, open-source licensing (GPL, MIT), and digital rights management (DRM); ethical hacking and bug bounty programs promote security.
知识产权问题涉及软件盗版、开源许可（GPL、MIT）和数字版权管理（DRM）；道德黑客和漏洞奖励计划促进安全。
The digital divide separates those with access to technology and skills from those without, exacerbating social and economic inequality; initiatives aim to provide affordable connectivity.
数字鸿沟将拥有技术和技能的人与没有的人分隔开来，加剧了社会和经济不平等；各项倡议旨在提供负担得起的连接。
Environmental impacts include e-waste from discarded electronics, energy consumption of data centres, and the potential for technology to monitor and reduce carbon footprints.
环境影响包括废弃电子产品产生的电子垃圾、数据中心的能源消耗，以及技术监测和减少碳足迹的潜力。
Professional codes of conduct, such as those from BCS, demand computing professionals act with integrity, respect confidentiality, and keep skills up to date.
职业道德准则（如 BCS 制定的规范）要求计算专业人员诚信行事、尊重机密并保持技能更新。
Published by TutorHao | Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导