Risk rating
A risk rating is a numerical or categorical assessment of the estimated level of potential loss, harm, or other undesirable outcomes associated with an entity or situation.
Inside AI
A risk score is a numerical estimate of the likelihood or severity of an undesirable event such as fraudulent activity.
- How it is used
- Uses risk scores to guide card authorization and inform criminal-justice decisions.
- How deep it goes
- Explains transaction thresholds and adaptive scoring, but provides no scoring formula or validation procedure.
- Examples
- Illustrates a seasonally unusual card purchase and 0–10 scores used in bail, sentencing, and parole decisions.
- What you can do after
- Evaluate/understand it by checking what a score draws on, what decision it affects, and whether that use risks false positives or bias.
Machine Learning System Design
A numerical value representing the estimated risk associated with a borrower or other subject.
- How it is used
- Frames borrower approval and rejection as decisions governed by estimated risk and the business consequences of mistakes.
- How deep it goes
- Defines binary and continuous outputs, then connects error costs to requirements and metrics without detailing scoring algorithms.
- Examples
- Uses loan approval, borrower default, successful repayment, and comparisons of people, credits, or money lent.
- What you can do after
- Evaluate risk-system requirements and metrics by comparing the financial consequences of false approvals and false rejections.
Read how Machine Learning System Design explains risk rating →
Data-Driven Cybersecurity
A risk score is a numerical or ordinal value representing the assessed severity or potential impact of a security risk.
- How it is used
- Guides investment, defense prioritization, and response decisions by translating vulnerabilities and threat conditions into comparable exposure measures.
- How deep it goes
- Explains composite risk factors and continuous updating, while deferring detailed index construction and statistical implementation.
- Examples
- Uses SecureShield’s dashboard comparison, adaptive scoring, anomaly detection, threat intelligence, and changes in detection and response metrics.
- What you can do after
- Evaluate current exposure, prioritize defenses, and adjust response efforts using changing risk scores and threat data.
Related concepts
- dynamic risk assessmentproduced by · A dynamic risk assessment is an evaluation that repeatedly updates risk estimates as relevant behavior, context, or operating conditions change.
- dashboardused by · A dashboard is a visual interface that consolidates selected data, metrics, or status information for monitoring, analysis, or decision-making.
- security scannerproduced by · A security scanner is a software tool that analyzes code, dependencies, configurations, systems, or artifacts to identify known or suspected vulnerabilities and other security weaknesses.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.