ICCP CDP flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Data Governance Definition

The exercise of authority and control over data management: decision rights, policies, standards, and accountability for data assets across the organization.

Data Steward

A business-side role accountable for data quality, definitions, and issue resolution within an assigned domain, executing the owner's decisions day to day.

Data Custodian

Typically an IT role responsible for physical storage, backup, access provisioning, and technical security controls for data, following owner-approved policy.

Business Glossary

A governed collection of business terms with approved definitions, synonyms, and accountable owners, used to align meaning across business and technical audiences.

Data Classification

Assigning data to sensitivity tiers such as public, internal, confidential, and restricted, which then determine handling, encryption, retention, and access requirements.

Data Owner

The accountable role for a data domain who approves access, funds quality work, and accepts risk on behalf of the business.

Policy Versus Standard

A policy states mandatory direction and intent; a standard specifies concrete, measurable requirements that implement the policy. Procedures describe step-by-step execution.

GDPR Data Minimization

A principle requiring collection of only personal data that is adequate, relevant, and limited to what is necessary for the stated purpose.

Metadata Types

Business metadata covers definitions and ownership; technical metadata covers schemas and structures; operational metadata covers job logs, access counts, and runtime statistics.

Master Versus Reference Data

Master data describes core business entities like customers and products; reference data is controlled code sets such as country or status codes used to classify other data.

Third Normal Form

A table is in 3NF when it is in 2NF and has no transitive dependency: non-key attributes depend only on the key, not on other non-key attributes.

Data Lineage

Documentation of the origin and transformation path of data from source to consumption, supporting impact analysis, root-cause investigation, and audit.

MDM Registry Style

A master data approach that builds an index of matched entity keys without changing source systems, providing a cross-reference for downstream lookup.

ACID Properties

Atomicity, consistency, isolation, and durability: guarantees that a transaction completes fully or not at all and leaves committed state intact.

Data Quality Dimensions

Measurable characteristics such as completeness, accuracy, consistency, timeliness, uniqueness, and validity, each requiring an explicit rule and threshold.

Data Lifecycle Stages

Create or acquire, maintain and use, archive, and purge. Retention and destruction rules attach to the record regardless of storage medium.

Fact Table Grain

The precise meaning of one fact row, such as one order line. Every measure stored in the fact table must be true at that level.

Star Schema

A dimensional design with one central fact table joined to denormalized dimension tables, minimizing joins and optimizing query performance.

SCD Type 2

A slowly changing dimension technique that inserts a new row per change with version, effective dates, and a current-row flag to preserve full history.

Semi-Additive Measure

A fact measure that can be summed across some dimensions but not time, such as account balance or inventory on hand.

Conformed Dimension

A dimension table shared across multiple fact tables with identical keys and attribute definitions, enabling consistent drill-across reporting.

Factless Fact Table

A fact table with no numeric measures, used to record event occurrences or coverage such as attendance or product-promotion assignment.

Inmon Versus Kimball

Inmon builds a normalized enterprise warehouse first, then derives marts; Kimball builds dimensional marts around conformed dimensions and bus matrix.

Additive Measure

A fact measure that can be summed across every dimension in the schema, such as sales quantity or revenue amount.

Precision and Recall

Precision is true positives over predicted positives; recall is true positives over actual positives. F1 is their harmonic mean.

Bias Versus Variance

High bias means the model underfits and misses structure; high variance means it overfits training noise and generalizes poorly to new data.

Cross-Validation

A resampling method that trains on k-1 folds and validates on the held-out fold, rotating k times and averaging performance estimates.

Data Leakage

A modeling error where information unavailable at prediction time, such as target-derived or future data, contaminates training and inflates reported performance.

P-Value Meaning

The probability of observing data at least as extreme as the sample, assuming the null hypothesis is true. It is not the probability the null is true.

Analytics Maturity Levels

Descriptive answers what happened, diagnostic explains why, predictive forecasts what will happen, and prescriptive recommends what action to take.

Type I and Type II Errors

Type I is a false positive, rejecting a true null hypothesis; Type II is a false negative, failing to reject a false null hypothesis.

Supervised Versus Unsupervised

Supervised learning trains on labeled examples to predict a target; unsupervised learning finds structure such as clusters or associations without labels.