Concept glossary
Put each term back in context. Every definition links to lessons where you can apply it.
- Scope 授權範圍
Assets, actions, time windows and limits established by an authorized owner; technical reachability does not extend permission.
- Evidence boundary 證據邊界
The separation of observation, inference and unknowns that keeps a claim within what the evidence supports.
- Model evaluation foundations: training, validation, testing and leakage
- CIA and security controls: define what needs protection
- Ethical hacking starts with authorization, scope and stop conditions
- Reconnaissance notes: separate observation from inference
- Port states: what do open, closed and filtered tell you?
- Enumeration needs service evidence: what follows an open 445?
- Vulnerability severity and remediation priority
- Malware triage: indicators, behavior and response evidence
- Packet visibility: what does encryption protect?
- Web server hardening: from exposure to verifiable configuration
- A web request’s journey: DNS, TCP, TLS and HTTP
- CIA triad CIA 三要素
Confidentiality limits unauthorized disclosure, integrity concerns improper change, and availability supports authorized use when needed.
- Hash 雜湊
A function mapping input bytes to a fixed-length digest; comparison still needs a trusted reference.
- DNS 網域名稱系統
A distributed resource-record system supporting name resolution; a mapping is not proof of ownership or permission.
- Port state 連接埠狀態
A tool’s description of a port response under a particular time, vantage, protocol and method.
- Enumeration 列舉
Identifying named resources or attributes from service responses while retaining identity and query conditions.
- Contextual risk 情境風險
A judgment combining threats, weaknesses, exposure and possible impact; a severity score alone is not a complete decision.
- Least privilege 最小權限
Grant only capabilities needed for the task and verify that disallowed actions are denied.
- Indicator 可疑指標
A clue for matching or investigation, such as a hash or process event, interpreted with source, time and context.
- Metadata 中繼資料
Information describing communication or data, such as timing and size, which is not automatically the application content.
- TLS 傳輸層安全
A protocol protecting communication between endpoints and supporting identity verification; it does not guarantee intent or content correctness.
- Independent verification 獨立確認
Checking identity and authority through an established trusted channel not controlled by the same suspicious source.
- Availability 可用性
Reliable access to services or data for authorized users when needed, considering capacity, recovery and operational constraints.
- Session 工作階段
A mechanism preserving interaction identity and state; identifiers require protection and designed invalidation.
- Firewall policy 防火牆政策
Rules allowing or denying traffic by conditions such as source, destination and protocol, tested with allowed and denied cases.
- Detection visibility 偵測可見性
The scope within which sensors and rules can observe, recognize and report events.
- Hardening 系統加固
Reducing unnecessary exposure and risk through minimal services, updates, permissions and verified configuration.
- Parameterized query 參數化查詢
A fixed query structure with inputs bound as values through the database interface so data does not change syntax.
- Wireless trust boundary 無線信任邊界
The distinction between network access, link protection, endpoint identity and application authorization.
- Mobile permission 行動應用權限
A platform control over restricted data or actions, checked against feature, timing and current grant state.
- Operational technology (OT) 操作技術
Systems monitoring or controlling physical processes, with security decisions accounting for performance, reliability and safety.
- Segmentation 網路分區
Constraining communication between asset zones according to necessary flows while preserving legitimate operations.
- Digital signature 數位簽章
Private-key signing and public-key verification supporting integrity and origin, dependent on trust in the key-to-identity relationship.
- Encryption 加密
A confidentiality mechanism allowing an appropriate key holder to recover data, alongside integrity protection and key management.
- Data leakage (model evaluation) 資料洩漏（模型評估）
Development or training uses information that should be unavailable for the evaluation, distorting the intended generalization question.
- Generalization 泛化
Model performance on new examples or situations within a declared scope, supported by suitable independent evaluation.
- HTTP HTTP
An application protocol expressing methods, resources, status and content through requests and responses.