📚 Year 13 CAIE Computer Science: Vocabulary & Terminology Memorization Guide | A-Level 计算机科学词汇术语速记指南
Mastering key terminology is essential for success in the CAIE A-Level Computer Science exam. This guide provides concise definitions and memory aids for the most important Year 13 terms, covering data representation, networks, hardware, security, algorithms, programming, databases, and more. Each term is presented with a twin explanation in English and Chinese, together with a quick tip to reinforce your revision.
掌握关键术语对于 CAIE A-Level 计算机科学考试至关重要。本指南为 Year 13 最重要的术语提供简明定义与记忆辅助，涵盖数据表示、网络、硬件、安全、算法、编程、数据库等领域。每个术语均以英文和中文的配对解释呈现，并附上速记窍门，帮助巩固复习。
1. Data Representation | 数据表示
Floating‑point representation stores a real number in the form ± mantissa × 2exponent. The mantissa holds the significant bits, while the exponent scales the value. Normalisation of a binary floating‑point number means adjusting the mantissa so that its most significant bit is 1 (for positive numbers), maximising precision. Memory aid: think of the mantissa as the “body” of the number and the exponent as a “ladder” that shifts the binary point.
浮点表示法以 ± 尾数 × 2指数 的形式存储实数。尾数保存有效位，指数决定数值的量级。对二进制浮点数进行规格化意味着调整尾数，使其最高有效位为 1（正数情况），从而让精度最大化。记忆技巧：把尾数想象成数字的“身体”，指数是移动小数点的“梯子”。
Precision vs range trade‑off: In a fixed‑size floating‑point word, allocating more bits to the mantissa increases precision, while more bits for the exponent extends the range of representable values. It is impossible to maximise both simultaneously. Tip: “More mantissa bits → sharper details; more exponent bits → bigger spread.”
精度与范围的权衡：在固定字长的浮点数中，给尾数分配更多位能提升精度，给指数更多位能扩大可表示值的范围。二者不可同时最大化。口诀：“尾数多→更精细，指数多→范围广”。
Unicode is a universal character encoding standard that assigns a unique code point to every character across all major writing systems. Unlike ASCII (limited to 128 or 256 characters), Unicode can represent over 143,000 characters using variable‑length encodings such as UTF‑8 and UTF‑16. Mnemonic: “Uni” = one – one unified scheme for the entire world.
Unicode（统一码）是一个通用字符编码标准，为全球所有主要书写系统的每个字符分配唯一的码点。与 ASCII（限于 128 或 256 个字符）不同，Unicode 可使用可变长编码（如 UTF‑8 和 UTF‑16）表示超过 14.3 万个字符。记忆法：“Uni” 意为“单一”——一套方案统管全世界。
Lossy vs lossless compression: Lossy compression permanently discards some data to reduce file size (e.g., MP3, JPEG), and the original cannot be perfectly reconstructed. Lossless compression retains every bit of the original data (e.g., ZIP, PNG), allowing exact restoration. Think: “Lossy loses information; lossless loses nothing.”
有损压缩与无损压缩：有损压缩通过永久丢弃部分数据来减小文件大小（如 MP3、JPEG），原始数据无法完全重建。无损压缩保留原始数据的每一位（如 ZIP、PNG），可以精确还原。联想：“有损牺牲质量，无损分毫不差”。
2. Communication & Internet Technologies | 通信与互联网技术
Packet switching: Data is split into small packets, each containing a header with source/destination addresses and sequence numbers. Packets may travel along different routes and are reassembled at the destination. Memory aid: imagine a book split into numbered pages, sent via different postmen, and reassembled on arrival.
分组交换：数据被拆分为小数据包，每个包含有包含源/目的地址和序号的包头。数据包可能沿不同路径传输，在目的地重新组装。记忆技巧：把一本书拆成编号的散页，通过不同邮递员发出，收到后再按号码装订。
TCP/IP protocol suite layers: Application (HTTP, FTP, SMTP), Transport (TCP, UDP), Internet (IP), and Link/Network Interface. Often remembered from top to bottom as “All People Try Internet Love” or simply by the acronym ATIL. The suite ensures reliable end‑to‑end communication over interconnected networks.
TCP/IP 协议族层次：应用层（HTTP、FTP、SMTP）、传输层（TCP、UDP）、互联网层（IP）和链路/网络接口层。常从上到下记为“应用传输网络链路”，可联想“ATIL”帮助记忆。该协议族保证互联网中端到端的可靠通信。
DNS (Domain Name System) translates human‑readable domain names (e.g., http://www.example.com) into IP addresses. It operates as a distributed database, avoiding the need to memorise numerical addresses. Analogous to a phone book for the internet.
域名系统 (DNS) 将人类可读的域名（如 http://www.example.com）转换为 IP 地址。它像分布式数据库一样运作，免去记忆数字地址的麻烦。好比互联网的电话号码簿。
IPv4 vs IPv6: IPv4 uses 32‑bit addresses, limiting the address space to about 4.3 billion; IPv6 uses 128‑bit addresses, providing an almost infinite pool. IPv6 also introduces simplified headers and built‑in security. Quick recall: “Four is small; six is big and secure.”
IPv4 与 IPv6：IPv4 采用 32 位地址，地址空间仅约 43 亿；IPv6 使用 128 位地址，地址池几乎无穷无尽。IPv6 还引入了更简化的头部和内置安全特性。速记：“4 小 6 大且安全”。
3. Hardware & Virtual Machines | 硬件与虚拟机
RISC vs CISC: RISC (Reduced Instruction Set Computer) processors use a small, highly optimised set of instructions, each executing in one clock cycle. CISC (Complex Instruction Set Computer) processors have a larger set of complex instructions that may take multiple cycles. RISC relies on software to combine simple operations, while CISC builds complexity into hardware. Memory aid: “RISC – Reduced, Rapid; CISC – Complicated, Cycle‑hungry.”
精简指令集 (RISC) 与复杂指令集 (CISC)：RISC 处理器使用一组精简且高度优化的指令，每条指令在一个时钟周期内执行。CISC 处理器拥有更多复杂指令，可能占用多个周期。RISC 依靠软件组合简单操作，CISC 则将复杂性固化在硬件里。记忆：“RISC精简快速，CISC复杂慢吞”。
Virtual machine: A software‑based emulation of a real computer, running an OS and applications as if they were on physical hardware. The hypervisor manages multiple VMs on a single host, isolating them from one another. Think of a virtual machine as a “computer inside a file”.
虚拟机：一种基于软件的物理计算机仿真，能够运行操作系统和应用程序，仿佛它们位于真实硬件之上。Hypervisor（虚拟机监视器）管理同一物理主机上的多台虚拟机，并使其彼此隔离。可把虚拟机想象为“文件里的计算机”。
Parallel processing – SIMD: Single Instruction, Multiple Data (SIMD) allows a single instruction to operate on several data elements simultaneously. It is commonly used in graphics and scientific computing. Tip: “One command, many data lanes.”
单指令多数据流 (SIMD)：一条指令可同时对多个数据元素执行相同操作，常用于图形和科学计算。口诀：“一条命令，多条数据通道并行跑”。
4. System Software & Operating Systems | 系统软件与操作系统
Interrupt handling: An interrupt is a signal sent to the processor indicating an event that needs immediate attention. The processor saves its current state, executes an Interrupt Service Routine (ISR), then restores the state. Mnemonic: “Stop, save, serve, resume.”
中断处理：中断是发送给处理器、表示需要立即关注事件的信号。处理器保存当前状态，执行中断服务程序 (ISR)，然后恢复状态。记忆法：“暂停、保存、服务、恢复”。
Scheduling algorithms: The OS uses scheduling to decide which process runs next. Round‑robin gives each process a fixed time slice (quantum); priority scheduling assigns ranks; shortest job first minimises waiting time. Recall: “Round‑robin shares evenly, priority jumps the queue.”
调度算法：操作系统通过调度决定下一个运行的进程。轮转法为每个进程分配固定时间片；优先级调度按等级分配；最短作业优先最小化等待时间。联想：“轮转均分，优先级插队”。
Linker and loader: A linker combines multiple object code files into a single executable, resolving external references. The loader then copies the executable from disk into main memory and prepares it for execution. Tip: “Linker glues pieces; loader launches.”
链接器与加载器：链接器将多个目标代码文件合并为一个可执行文件，并解析外部引用。加载器随后把可执行文件从磁盘复制到内存，为运行做好准备。口诀：“链接器粘合碎片，加载器点火发射”。
5. Security, Privacy & Data Integrity | 安全、隐私与数据完整性
Symmetric encryption uses a single shared key for both encryption and decryption (e.g., AES). It is fast but requires secure key distribution. Asymmetric encryption uses a public/private key pair: the public key encrypts, and only the matching private key decrypts (e.g., RSA). Think: “Symmetric = one key, both lock/unlock; Asymmetric = two keys, like a mailbox and a key.”
对称加密使用同一个共享密钥进行加密和解密（如 AES），速度快但需要安全分发密钥。非对称加密使用公钥/私钥对：公钥加密，仅对应的私钥解密（如 RSA）。联想：“对称加密一把钥匙开关锁；非对称加密两把钥匙——一个投信口，一个开锁”。
Digital signature: A hash of the message is encrypted with the sender’s private key. The recipient decrypts it with the sender’s public key and compares the hash to verify integrity and authenticity. It confirms the message has not been tampered with. Memory aid: “Sign with the private key – nobody else can; verify with the public key – anyone can check.”
数字签名：使用发送者的私钥对消息的哈希值进行加密。接收方用发送者的公钥解密哈希，并与重新计算的哈希比对，以验证完整性和真实性。这能确认消息未被篡改。记忆：“私钥签名，独一无二；公钥验签，人人可查”。
Firewall: A hardware or software barrier that filters incoming and outgoing network traffic based on predefined security rules. It inspects packets and blocks unauthorised access. Think of it as a security guard checking IDs at a building entrance.
防火墙：基于预设安全规则过滤进出网络流量的硬件或软件屏障。它检查数据包并拦截未授权的访问。好比大厦入口查验身份证件的保安。
SQL injection: A code injection technique that exploits vulnerabilities in an application’s database layer, typically by inserting malicious SQL statements into input fields (e.g., ‘ OR 1=1 –). Always use parameterised queries to prevent it. Warning: “Never trust user input.”
SQL 注入：一种利用应用程序数据库层漏洞的代码注入技术，通常是在输入字段中插入恶意 SQL 语句（例如 ‘ OR 1=1 –）。务必使用参数化查询加以防范。警示：“永远别相信用户输入”。
6. Ethics & Ownership | 伦理与所有权
Open‑source software is distributed with its source code, allowing anyone to study, modify, and share it (e.g., Linux). Proprietary software keeps its source code secret and restricts user rights (e.g., Microsoft Windows). Quick contrast: “Open = free to view and adapt; Proprietary = locked and licensed.”
开源软件随附源代码分发，允许任何人学习、修改和分享（如 Linux）。专有软件将源代码保密，并限制用户权利（如 Microsoft Windows）。快速对比：“开源自由查看改造，专有上锁付费授权”。
Data privacy and the GDPR: Regulations like the General Data Protection Regulation require organisations to protect personal data, obtain consent, and allow individuals to access or delete their information. The core principle: “You own your data, not the company.”
数据隐私与 GDPR：诸如《通用数据保护条例》之类的法规要求组织保护个人数据、获取同意，并允许个人访问或删除自身信息。核心原则：“数据属于你，而非公司”。
AI ethics: Concerns around artificial intelligence include bias in training data, lack of transparency (black‑box decisions), and accountability. Developers must ensure fairness and explainability. Memory hook: “Algorithms are mirrors of their data – if the data is biased, the AI will be too.”
人工智能伦理：围绕人工智能的关切包括训练数据中的偏见、缺乏透明性（黑箱决策）以及问责制。开发者必须确保公平性和可解释性。记忆点：“算法是其数据的镜子——数据有偏见，AI 就会学歪”。
7. Algorithm Design & Problem Solving | 算法设计与问题求解
Recursion: A function calls itself with a smaller instance of the same problem until it reaches a base case. The call stack keeps track of pending operations. Classic example: factorial n! = n × (n−1)!. Tip: “Recursive thinking: solve a tiny piece, let the function handle the rest.”
递归：函数调用自身来求解同一问题的更小实例，直到触及基准情形。调用栈负责跟踪未完成的操作。经典例子：阶乘 n! = n × (n−1)!。窍门：“递归思维：先解决一小块，剩下的交给自己”。
Big O notation describes the worst‑case time or space complexity of an algorithm. O(1) is constant time, O(log n) is logarithmic, O(n) is linear, O(n²) is quadratic. It estimates how the runtime grows with input size. Mnemonic: “Big O = Oh‑my‑growth!” and “O(1) fastest, O(n²) beware.”
大 O 表示法描述算法最坏情况下的时间或空间复杂度。O(1) 常量时间，O(log n) 对数级，O(n) 线性，O(n²) 平方级。它评估运行时间随输入规模增长的速率。记忆：“Big O 看增长率，O(1) 闪电快，O(n²) 要当心”。
Abstract Data Types (ADTs) – Stack and Queue: A stack follows Last‑In‑First‑Out (LIFO) like a pile of plates; a queue follows First‑In‑First‑Out (FIFO) like a waiting line. They define logical behaviour, not implementation. Recall: “Stack – push/pop vertically; Queue – enqueue/dequeue horizontally.”
抽象数据类型 (ADT) —— 栈与队列：栈遵循后进先出 (LIFO)，如同一叠盘子；队列遵循先进先出 (FIFO)，仿佛排队等候。它们定义的是逻辑行为而非具体实现。记忆：“栈上下操作，队列横向流动”。
8. Programming Paradigms & Translation | 编程范式与翻译
Encapsulation bundles data (attributes) and methods that operate on that data within a class, hiding the internal state. Access is controlled via public, private, and protected modifiers. Think of a capsule that protects its contents.
封装将数据（属性）和操作这些数据的方法捆绑在一个类里，隐藏内部状态。通过 public、private、protected 修饰符控制访问。联想胶囊保护内部物质。
Inheritance allows a child class to derive properties and behaviours from a parent class, promoting code reuse. The child can override methods to provide specialised functionality. Memory aid: “Child inherits family traits but can develop its own personality.”
继承允许子类从父类派生属性和行为，促进代码重用。子类可重写方法以提供特定功能。记忆法：“孩子继承家族特征，但也能发展自己的个性”。
Compiler vs Interpreter: A compiler translates the entire source code into machine code before execution (e.g., C++). An interpreter translates and executes line‑by‑line (e.g., Python). Compilers produce standalone executables; interpreters need to reside in memory. Just‑in‑time (JIT) compilation combines both. Recall: “Compiler cooks the whole meal; interpreter serves bite‑by‑bite.”
编译器与解释器：编译器在执行前将全部源代码翻译为机器码（如 C++）。解释器逐行翻译并执行（如 Python）。编译器生成独立的可执行文件；解释器则需常驻内存。即时编译 (JIT) 结合二者。联想：“编译器一顿做完全席；解释器一口一口喂”。
Polymorphism allows objects of different classes to respond to the same method call in a way appropriate to their own type. Method overriding and overloading are common forms. Simple image: “Many forms, one interface.”
多态允许不同类的对象用各自合适的方式响应相同的方法调用。方法重写和重载是常见形式。简单画面：“多种形态，同一接口”。
9. Data Structures & ADTs | 数据结构与抽象数据类型
Linked list: A dynamic data structure where each node contains data and a pointer to the next node. It allows efficient insertion and deletion without shifting elements, unlike arrays
Published by TutorHao | Year 13 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