Domain knowledge
Domain knowledge is specialized understanding of the concepts, facts, practices, relationships, and constraints of a particular subject area.
Experimentation for Engineers
Domain knowledge is specialized knowledge about the application area, processes, and constraints relevant to a system.
- How it is used
- Constrains experimental designs and supplements decisions about system changes before business-metric impact is measured.
- How deep it goes
- Defines domain knowledge, explains its predictive limits, and works through a parameter-validity constraint without formalizing acquisition methods.
- Examples
- Uses a proprietary trading strategy whose nonnegative signal threshold prevents overlapping buy and sell decisions.
- What you can do after
- Evaluate/understand when business or engineering knowledge can safely restrict a design space but cannot replace an experiment.
Read how Experimentation for Engineers explains domain knowledge →
Inside AI
Domain knowledge is specialized understanding of the facts, practices, constraints, and terminology of a particular subject area.
- How it is used
- Uses domain knowledge to validate AI conclusions against the realities of the problem being modeled.
- How deep it goes
- Explains its role through a model failure and clinical reasoning, without giving a formal validation procedure.
- Examples
- Illustrates it with a misleading asthma-and-pneumonia risk pattern and the information-gathering work of a medical visit.
- What you can do after
- Diagnose/debug with it by involving domain experts to question model patterns and check expected responses against subject-matter realities.
AI-Powered Search
Domain-specific knowledge is information, concepts, terminology, and relationships pertaining to a particular subject area.
- How it is used
- Interprets domain-specific queries and documents so search can move beyond keyword matching toward user intent.
- How deep it goes
- Describes representations and query and batch-processing roles, but provides no domain-model schema or implementation code.
- Examples
- Illustrates structured queries on tags, categories, or entities and spelling corrections informed by user choices.
- What you can do after
- Evaluate/understand it by distinguishing keyword matching from domain-aware semantic search and recognizing where query-time domain intelligence is needed.
The Well-Grounded Data Analyst
Domain expertise is specialized knowledge of the subject area represented by the data or analysis.
- How it is used
- Uses subject-area understanding to interpret addresses, challenge automated categorization, and communicate meaningful findings to stakeholders.
- How deep it goes
- Demonstrates iterative validation and manual correction, but stops short of formalizing domain-expert elicitation or reusable geographic rules.
- Examples
- Uses London addresses, “Other” records, the Hull abbreviation, city-list comparison, and spending by city.
- What you can do after
- Diagnose misleading categorical results and improve them through targeted inspection, validation, and correction.
Read how The Well-Grounded Data Analyst explains domain knowledge →
Causal Inference for Data Science
Domain knowledge is knowledge about the real-world subject matter, mechanisms, and context represented by a dataset or model.
- How it is used
- Guides causal inference by translating real-world mechanisms into graphs, assumptions, and defensible adjustment choices.
- How deep it goes
- Explains causal direction, confounder selection, graph construction, positivity, and adjustment while emphasizing limitations from missing variables.
- Examples
- Uses kidney-stone treatment assignment, causal graphs, propensity-score variables, and comparisons between predictive accuracy and causal validity.
- What you can do after
- Evaluate causal questions, collaborate with domain experts, identify confounders, and select adjustments supported by the causal structure.
Read how Causal Inference for Data Science explains domain knowledge →
Starting Data Analytics with Generative AI and Python
Subject-matter understanding of the meanings, expected values, and valid relationships in an application's domain.
- How it is used
- Guides research-question design, analytical interpretation, data-quality review, and oversight of AI-generated methods.
- How deep it goes
- Connects expected patterns, business meaning, outlier decisions, and AI review, but does not provide a domain-specific decision template.
- Examples
- Uses expected chart shapes, correlation meaning, a New York–Los Angeles transaction anomaly, and value_counts() for review.
- What you can do after
- Evaluate findings, validate anomalies, refine AI analysis, and test generated code against business expectations.
Read how Starting Data Analytics with Generative AI and Python explains domain knowledge →
Data Without Labels
Domain knowledge is subject-matter understanding about the real-world context represented by data and analytical results.
- How it is used
- Checks whether patterns found in data support credible business conclusions and actions.
- How deep it goes
- Explains the need to test a rule against business logic, but gives no formal procedure for acquiring domain knowledge.
- Examples
- Contrasts shopping-basket rules about eggs, bread, cheese, and milk using a five-invoice transaction table.
- What you can do after
- Evaluate/understand whether an association rule makes business sense before using it to guide a decision.
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
Domain knowledge is expertise about the subject matter and phenomena represented by a dataset.
- How it is used
- Guides feature engineering, data cleaning, feature selection, exploratory analysis, validation, and model preparation.
- How deep it goes
- Explains practical decisions with real-estate and collinearity examples but stops short of a general method for acquiring domain expertise.
- Examples
- Uses car-fuel relationships, duplicated numeric features, and real-estate bathrooms, parking, furnishing, property type, and location fields.
- What you can do after
- Apply it to design meaningful features, select among redundant predictors, justify imputations, and adapt modeling workflows to the problem.
AI Agents in Action, Second Edition
- How it is used
- Supplies agents with retrieved, task-specific context beyond the model’s initially available information.
- How deep it goes
- Defines retrieval and augmentation, compares storage and search routes, and gives implementation guidance without detailing model internals.
- Examples
- Compares PDFs and database tables in a retrieval table and filters a vector knowledge base by product, version, role, and tenant.
- What you can do after
- Operate/configure it through selective retrieval, metadata filters, hybrid search, and answers grounded in retrieved passages.
Read how AI Agents in Action, Second Edition explains domain knowledge →
Outlier Detection in Python
Domain knowledge is application-specific knowledge about entities, processes, relationships, and conditions relevant to data analysis.
- How it is used
- Guides feature-subspace selection when detecting meaningful combinations of values in outlier detection.
- How deep it goes
- Explains the selection rationale and computational trade-offs, but does not provide a domain-specific selection checklist.
- Examples
- Uses staff-expense fields and baseball-statistic feature sets as concrete feature-selection artifacts.
- What you can do after
- Evaluate which feature combinations may produce interpretable outliers while recognizing coverage and computation trade-offs.
Read how Outlier Detection in Python explains domain knowledge →
Coding with AI
Domain-specific knowledge is specialized understanding of the problem area to which software is applied.
- How it is used
- Guides developers in evaluating AI suggestions and adapting generated code to project-specific architecture, behavior, and user needs.
- How deep it goes
- Applies domain knowledge through evaluation, revision, verification, and edge-case handling rather than formalizing the knowledge itself.
- Examples
- Uses a Flask greeting application, session-ending JavaScript, question selection, database state, and duplicate-entry fallback handling.
- What you can do after
- Evaluate, revise, and verify AI-generated application code against architectural goals, expected workflows, state requirements, and edge cases.
The Art of AI Product Development
Domain knowledge is specialized understanding of the concepts, practices, workflows, and constraints of a particular field or industry.
- How it is used
- Guides specialized AI product choices, user workflows, evaluation criteria, and collaboration with stakeholders.
- How deep it goes
- Connects domain knowledge to datasets, benchmarks, user feedback, and model updates, but does not detail implementation algorithms.
- Examples
- Uses supply forecasting errors involving weather and supplier delays, plus feedback buttons, comments, roundtables, and roadmap input.
- What you can do after
- Apply domain knowledge to build relevant evaluations, refine AI behavior, and improve workflows through structured user feedback.
Read how The Art of AI Product Development explains domain knowledge →
Think Like a CTO
Domain knowledge is specialized understanding of a particular industry, business area, or subject matter.
- How it is used
- Applies expertise to board guidance, vendor selection, team capability, and awareness of industry changes.
- How deep it goes
- Evaluates expertise through relevant skills, practical exercises, accessibility, scalability, and references rather than price alone.
- Examples
- Uses board members’ guidance, release-team tools such as Jenkins and SonarQube, and vendor exercises as examples.
- What you can do after
- Evaluate vendor expertise through skills alignment and exercises, and prepare board presentations around members’ business focus.
AI for Everyday IT
Domain knowledge is specialized information and understanding about a particular subject area, profession, or operational context.
- How it is used
- Guides AI problem formulation by helping practitioners define the right questions, requirements, and boundaries.
- How deep it goes
- Explains its value in problem formulation but does not provide a domain-specific workflow.
- Examples
- Uses technical diagnosis, marketing strategy, and a paint-by-numbers analogy to illustrate structured AI guidance.
- What you can do after
- Apply domain knowledge through problem formulation to guide AI toward more effective professional outputs.
Knowledge Graphs and LLMs in Action
Specialized knowledge and contextual understanding of a particular scientific or professional domain.
- How it is used
- Grounds intelligent systems, graph analytics, and LLM outputs in specialized scientific or professional context.
- How deep it goes
- Applies expertise to interpretation, feature selection, validation, and knowledge-base refinement without detailing formal expertise elicitation.
- Examples
- Uses celiac-disease pathway analysis, DWPC feature engineering, established ontologies, and expert-refined knowledge bases.
- What you can do after
- Apply it to select interpretable features, validate analytical results, and produce contextually grounded recommendations.
Read how Knowledge Graphs and LLMs in Action explains domain knowledge →
The Complete Obsolete Guide to Generative AI
Specialized understanding and expertise concerning a particular subject or professional field.
- How it is used
- Guides prompting, reviewing, and applying LLMs to specialized documents and professional tasks.
- How deep it goes
- Explains its practical role through contract drafting and warns that generated details still require human checking.
- Examples
- Uses contract drafting, consulting terms, revenue shares, and specialized troubleshooting as concrete examples.
- What you can do after
- Evaluate generated specialized documents against known facts, rights, and risks instead of relying on fluent output automatically.
Read how The Complete Obsolete Guide to Generative AI explains domain knowledge →
Financial AI in Practice
Subject-matter understanding that guides the interpretation and transformation of data into meaningful variables.
- How it is used
- Guides financial feature engineering in the data asset layer, connecting raw data transformations to meaningful predictive variables.
- How deep it goes
- Defines its role and gives feature examples, but stops short of implementation steps or code.
- Examples
- Uses volatility measures, fraud-velocity variables, and seasonal consumer-spending patterns as feature examples.
- What you can do after
- Apply it to shape financial features that represent meaningful concepts rather than arbitrary correlations.
Read how Financial AI in Practice explains domain knowledge →
Related concepts
- informationproduced by · Information is content that conveys meaningful distinctions or reduces uncertainty about possible states, observations, or outcomes.
- retrieval-augmented generationused by · Retrieval-augmented generation is a technique that retrieves relevant information from external sources and supplies it as context to a generative model when producing an output.
- wisdomproduces · Wisdom is knowledge and insight applied with sound judgment and ethical consideration to guide decisions and actions.
- DIKW hierarchypart of · The DIKW hierarchy is a conceptual framework organizing data, information, knowledge, and wisdom as progressively interpreted forms of understanding.
- experimentationcontrasts with · Experimentation is the deliberate testing of an idea, intervention, or alternative by observing outcomes to gather evidence or compare effects.
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.