📚 Artificial Intelligence Revision Notes for IB and WJEC Computer Science | IB与WJEC计算机科学人工智能考点精讲
Artificial Intelligence (AI) is a core topic in both the IB Computer Science syllabus and the WJEC Computer Science specification. Understanding AI involves grasping how machines can simulate human cognitive functions such as learning, reasoning, and self-correction. This revision guide brings together key concepts, examination techniques, and essential terminology to help you consolidate your knowledge and perform confidently in assessments.
人工智能 (AI) 是 IB 和 WJEC 计算机科学课程的核心主题。理解 AI 需要掌握机器如何模拟人类认知功能，如学习、推理和自我纠正。本复习指南整合了关键概念、应试技巧和必备术语，帮助你巩固知识并在考试中自信发挥。
1. Defining Artificial Intelligence | 人工智能的定义
AI is the field of computer science that aims to create machines capable of performing tasks that normally require human intelligence. These tasks include reasoning, knowledge representation, planning, learning, natural language processing, perception, and the ability to move and manipulate objects.
人工智能是计算机科学的一个分支，旨在创造能够执行通常需要人类智能才能完成的任务的机器。这些任务包括推理、知识表示、规划、学习、自然语言处理、感知，以及移动和操作物体的能力。
A key distinction is between ‘narrow AI’, which is designed for a specific task, and ‘general AI’, which would possess the ability to understand, learn, and apply knowledge across a broad range of tasks at a human level.
一个关键区别在于“狭义人工智能”（针对特定任务设计）和“通用人工智能”（具备在广泛任务中像人类一样理解、学习和应用知识的能力）之间的差异。
In IB and WJEC exams, you are often asked to explain what AI is in simple terms and give examples of current narrow AI systems such as voice assistants and recommendation engines.
在 IB 和 WJEC 考试中，你常被要求用简单的语言解释什么是 AI，并给出当前狭义人工智能系统的例子，如语音助手和推荐引擎。
2. The Turing Test and Measuring Intelligence | 图灵测试与智能衡量
Proposed by Alan Turing in 1950, the Turing Test evaluates a machine’s ability to exhibit intelligent behaviour indistinguishable from that of a human. A human interrogator asks questions to both a machine and a human via a text interface; if the interrogator cannot reliably tell which is the machine, the machine is said to have passed the test.
图灵测试由艾伦·图灵于 1950 年提出，用于评估机器表现出与人类难以区分的智能行为的能力。人类提问者通过文本界面同时向机器和人类提问；如果提问者无法可靠分辨哪一个是机器，则称机器通过了测试。
Critics argue that passing the Turing Test does not equate to genuine understanding or consciousness. The Chinese Room thought experiment by John Searle challenges the idea that syntactic manipulation of symbols amounts to true intelligence.
批评者认为，通过图灵测试并不等同于真正的理解或意识。约翰·塞尔提出的“中文房间”思想实验挑战了符号的句法操作等同于真正智能的观点。
Exam questions may require you to describe the Turing Test and discuss its limitations, or compare it with other intelligence measures such as the Lovelace Test.
考试问题可能要求你描述图灵测试并讨论其局限性，或将其与其他智能衡量标准（如洛夫莱斯测试）进行比较。
3. Strong AI vs Weak AI | 强人工智能与弱人工智能
Weak AI (also known as narrow AI) refers to systems that are designed and trained for a specific task. They simulate human intelligence but do not possess consciousness. Most current AI applications, such as image recognition and spam filters, fall into this category.
弱人工智能（也称狭义人工智能）指为特定任务而设计和训练的系统。它们模拟人类智能，但不具备意识。当前大多数人工智能应用，如图像识别和垃圾邮件过滤器，都属于这一范畴。
Strong AI (or artificial general intelligence) would be a machine with the ability to apply intelligence to any problem, rather than just a specific domain, and would have self-awareness and consciousness. This remains theoretical and has not been achieved.
强人工智能（或通用人工智能）将是一种能够将智能应用于任何问题（而非仅仅特定领域），并具备自我意识和意识的机器。这仍然停留在理论层面，尚未实现。
The distinction is frequently examined: students should be able to classify a given AI system as weak or strong and justify their reasoning.
这一区别经常被考查：学生应能判断给定的 AI 系统属于弱智能还是强智能，并说明理由。
4. Expert Systems | 专家系统
An expert system is an early form of AI that emulates the decision-making ability of a human expert. It consists of a knowledge base, an inference engine, and a user interface. The knowledge base contains domain-specific facts and rules, while the inference engine applies logical rules to deduce new information or make recommendations.
专家系统是人工智能的早期形式，模拟人类专家的决策能力。它由知识库、推理引擎和用户界面组成。知识库包含领域特定的事实和规则，推理引擎应用逻辑规则来推导新信息或提供建议。
Common examples include medical diagnosis systems and fault-finding in electronic equipment. In WJEC, expert systems are a core topic, and questions often ask about the structure: knowledge base, rule base, inference engine, and how forward and backward chaining work.
常见的例子包括医疗诊断系统和电子设备故障排查。WJEC 课程中，专家系统是核心主题，问题常涉及结构：知识库、规则库、推理引擎，以及正向链和逆向链如何工作。
Forward chaining starts with known facts and applies rules to reach a conclusion, while backward chaining starts with a hypothesis and works backwards to find supporting evidence.
正向链从已知事实出发，应用规则得出结论；逆向链则从假设出发，反向寻找支持证据。
5. Knowledge Representation | 知识表示
Knowledge representation is the way in which AI systems store and organise information so that it can be used for reasoning. Semantic networks, frames, and production rules are common methods.
知识表示是人工智能系统存储和组织信息以便用于推理的方式。语义网络、框架和产生式规则是常用方法。
A semantic network represents knowledge as interconnected nodes and edges, where nodes represent concepts and edges denote relationships such as ‘is-a’ or ‘has-a’. Frames are data structures that group attributes and values, similar to objects in OOP. Production rules use if-then statements to encode conditional knowledge.
语义网络用互联的节点和边表示知识，其中节点表示概念，边表示“是一种”或“有”等关系。框架是分组属性和值的数据结构，类似于面向对象编程中的对象。产生式规则使用 if-then 语句编码条件知识。
IB learners should be able to compare these approaches and discuss their strengths for different AI tasks.
IB 学生应能比较这些方法，并讨论它们在不同 AI 任务中的优势。
6. Search Strategies and Problem Solving | 搜索策略与问题求解
Many AI problems are framed as search problems, where the solution is a path from an initial state to a goal state. Uninformed search algorithms, such as breadth-first search and depth-first search, explore the state space without additional knowledge about the goal’s location.
许多 AI 问题被定义为搜索问题，解决方案是从初始状态到目标状态的路径。无信息搜索算法，如宽度优先搜索和深度优先搜索，在没有关于目标位置的额外知识的情况下探索状态空间。
Informed or heuristic search uses a function to estimate the cost to reach the goal, guiding the search more efficiently. A common heuristic is the evaluation function f(n) = g(n) + h(n), where g(n) is the cost so far and h(n) is the heuristic estimate to the goal.
有信息或启发式搜索使用一个函数来估算到达目标的代价，从而更高效地引导搜索。常用的估价函数为 f(n) = g(n) + h(n)，其中 g(n) 为已消耗代价，h(n) 为到目标的启发式估值。
f(n) = g(n) + h(n)
WJEC exams often present a simple maze or graph and ask students to show how breadth-first, depth-first, or A* search would expand nodes.
WJEC 考试常给出一个简单的迷宫或图，要求学生展示宽度优先、深度优先或 A* 搜索如何扩展节点。
7. Machine Learning and Neural Networks | 机器学习与神经网络
Machine learning is a subset of AI where systems learn from data without being explicitly programmed. Supervised learning uses labelled data to train models, while unsupervised learning finds hidden patterns in unlabelled data. Reinforcement learning involves an agent learning through trial and error by receiving rewards or penalties.
机器学习是 AI 的一个子集，系统从数据中学习而无需显式编程。监督学习使用带标签的数据训练模型，无监督学习在无标签数据中发现隐藏模式。强化学习则涉及智能体通过试错并根据奖励或惩罚进行学习。
Artificial neural networks are inspired by the human brain and consist of interconnected layers of nodes (neurones). Each connection has a weight, and the network is trained by adjusting these weights through backpropagation.
人工神经网络受人类大脑启发，由相互连接的节点（神经元）层组成。每条连接都有一个权重，网络通过反向传播调整这些权重来进行训练。
In IB, you may be asked to explain basic concepts like training data, overfitting, and generalisation. WJEC may include simple perceptron calculations or identification of supervised vs unsupervised tasks.
在 IB 中，你可能需要解释基本概念，如训练数据、过拟合和泛化。WJEC 可能包括简单的感知机计算，或识别监督任务与无监督任务。
8. Natural Language Processing and Chatbots | 自然语言处理与聊天机器人
NLP enables computers to understand, interpret, and generate human language. Common subtasks include tokenisation, part-of-speech tagging, sentiment analysis, and machine translation.
自然语言处理使计算机能够理解、解释和生成人类语言。常见子任务包括分词、词性标注、情感分析和机器翻译。
Chatbots rely on NLP and often follow predefined scripts or AI models to simulate conversation. Rule-based chatbots use keyword matching and decision trees, while AI-based chatbots use machine learning to understand context and intent.
聊天机器人依赖自然语言处理，通常遵循预定义脚本或 AI 模型来模拟对话。基于规则的聊天机器人使用关键词匹配和决策树，而基于 AI 的聊天机器人使用机器学习理解上下文和意图。
Exam questions might ask you to discuss the ethical issues of chatbots or evaluate the limitations of NLP, such as ambiguity and cultural nuances.
考试问题可能会要求你讨论聊天机器人的伦理问题，或评估自然语言处理的局限性，如歧义和文化细微差别。
9. Perception and Robotics | 感知与机器人学
Perception involves allowing a machine to sense and interpret its environment using sensors such as cameras, lidar, and microphones. Computer vision is a key area where algorithms process images to recognise objects, faces, and movements.
感知涉及让机器使用摄像头、激光雷达和麦克风等传感器感知并解读其环境。计算机视觉是其中的一个关键领域，算法处理图像以识别物体、人脸和运动。
Robotics combines AI with mechanical engineering to create machines that can perform physical tasks autonomously. Drones, autonomous vehicles, and manufacturing robots are examples where AI-driven perception and control algorithms play a crucial role.
机器人学将人工智能与机械工程相结合，创造出能够自主执行物理任务的机器。无人机、自动驾驶汽车和制造机器人就是 AI 驱动的感知与控制算法发挥关键作用的例子。
Both IB and WJEC might include questions on sensors and actuators, and how a robot can use feedback loops to adjust its behaviour in real time.
IB 和 WJEC 都可能包含关于传感器与执行器，以及机器人如何使用反馈回路实时调整行为的问题。
10. AI Ethics, Bias, and Social Impact | AI 伦理、偏见与社会影响
The rapid development of AI raises significant ethical concerns. Bias in training data can lead to unfair outcomes in areas like hiring, policing, and loan approvals. Algorithmic fairness, transparency, and accountability are essential to ensure responsible AI deployment.
人工智能的快速发展引发了重大的伦理关切。训练数据中的偏见可能导致在招聘、警务和贷款审批等领域产生不公平的结果。算法公平、透明和问责对于确保负责任的 AI 部署至关重要。
Privacy is another key issue, as AI systems often process vast amounts of personal data. Students should be able to discuss the balance between data utility and privacy, referencing regulations like GDPR.
隐私是另一个关键问题，因为 AI 系统常常处理海量个人数据。学生应能讨论数据效用与隐私之间的平衡，并引用 GDPR 等法规。
Exam questions frequently ask for arguments both for and against AI technologies, such as autonomous weapons or mass surveillance, and require a balanced conclusion.
考试问题常要求学生分别提出支持和反对 AI 技术的论点，如自主武器或大规模监控，并给出平衡的结论。
11. Common AI Application Scenarios | 常见 AI 应用场景
AI is embedded in everyday technology: search engines, email spam filters, voice assistants (Siri, Alexa), recommendation systems (Netflix, Amazon), and fraud detection in banking.
人工智能已融入日常技术：搜索引擎、邮件垃圾过滤器、语音助手（Siri、Alexa）、推荐系统（Netflix、Amazon）以及银行欺诈检测。
In healthcare, AI assists in analysing medical images, predicting patient outcomes, and personalising treatment plans. In education, adaptive learning systems tailor content to individual student needs.
在医疗领域，AI 辅助分析医学影像、预测患者预后并制定个性化治疗方案。在教育领域，自适应学习系统根据学生个人需求定制内容。
When answering application-based questions, it is useful to reference appropriate AI techniques: for instance, a recommendation system may use collaborative filtering (a type of machine learning).
回答应用题时，引用恰当的 AI 技术很有帮助：例如，推荐系统可能采用协同过滤（一种机器学习）。
12. Exam Tips and Key Terminology | 应考提示与关键术语
For both IB and WJEC papers, focus on precise definitions and clear examples. Memorise the structure of an expert system, the difference between supervised and unsupervised learning, and the Turing Test criteria.
无论是 IB 还是 WJEC 试卷，都要注重准确定义和清晰的例子。记住专家系统的结构、监督与无监督学习的区别，以及图灵测试的评判标准。
Use technical vocabulary appropriately: knowledge base, inference engine, heuristic, neural network, training data, overfitting, weak vs strong AI. Practice drawing and explaining search trees and simple neural network diagrams.
恰当地使用技术词汇：知识库、推理引擎、启发式、神经网络、训练数据、过拟合、弱 AI 与强 AI。练习绘制并解释搜索树和简单的神经网络图。
| Term (English) | 中文术语 | Brief Definition |
|---|---|---|
| Weak AI | 弱人工智能 | AI designed for a specific task, no consciousness. |
| Inference Engine | 推理引擎 | Part of an expert system that applies rules to derive answers. |
| Heuristic | 启发式 | A ‘rule of thumb’ used to guide search or decision-making. |
| Turing Test | 图灵测试 | A test to see if a machine can exhibit intelligent behaviour indistinguishable from a human. |
Time management is crucial: allocate more marks to longer-answer questions that require structured arguments, such as ethical discussions or designing a simple AI solution.
时间管理至关重要：为需要结构化论证的长答题（如伦理讨论或设计简单的 AI 解决方案）分配更多分数对应的作答时间。
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