Google Advanced Data Analytics key terms

The relationship between an analysis result and the actual stakeholder choice, including population, outcome, uncertainty, cost and authorized responsibility.

A defined reference measurement for the full relevant task or service, not only one convenient step whose improvement can hide downstream costs.

The number of result rows produced by matching relationships between inputs, which can multiply totals when an assumed unique key has duplicates.

The library-defined matching of labels or positions during an operation, whose meaning must be checked rather than inferred from visual row order.

A justified rule for handling absent observations under the actual data and decision context, not an automatic substitution of every missing field with zero.

A probability of results at least as extreme under a specified null model and its assumptions, not the probability that the null hypothesis is true.

The long-run proportion of intervals constructed by a procedure that contain the parameter under its model, distinct from a posterior probability of one fixed parameter.

A fitted conditional relationship parameter under the stated model and included variables, which is not automatically the causal effect of an intervention.

Use of information across an intended training/evaluation boundary that can make a reported assessment less independent or more optimistic than claimed.

A database that stores data in tables made of rows and columns and uses relationships between tables. It is queried with languages such as SQL and follows the relational model.

A column or set of columns whose values uniquely identify each row in a database table. It cannot contain duplicate or null values.

A column or set of columns in one table that refers to the primary key of another table. It enforces referential integrity between related records.

Structured Query Language, a standard language for defining, querying, updating, and managing data in relational databases. It includes statements for creating tables and retrieving rows.

The process of organizing relational database tables to reduce duplicated data and avoid update anomalies. It typically involves dividing data into related tables and defining keys.

A data structure that speeds up searches on a database table by providing quick lookup paths to rows. It uses extra storage and must be updated when data changes.

Data that describes other data, such as a file's author, creation date, format, or size. In databases, it describes table structures, constraints, and other schema information.

The principle that users and systems should be granted only the minimum access rights necessary to perform their tasks, reducing the potential damage from accidents or malicious activity.

A control that divides critical tasks among different individuals so that no single person can complete a sensitive action alone, helping to prevent fraud and errors.

An access control model where permissions are assigned to roles rather than individual users, and users are assigned to roles based on their job responsibilities.

An authentication method that requires two or more independent credentials from different categories (something you know, have, or are) to verify a user's identity.

The process of verifying the identity of a user, device, or system, typically through credentials such as passwords, tokens, or biometrics.

The process of determining what actions or resources an authenticated entity is permitted to access, often enforced through access control policies.

A mechanism that allows a user to log in once and access multiple applications or systems without re-entering credentials, improving usability and centralizing access control.

The process of converting plaintext into ciphertext using an algorithm and a key, so that only authorized parties can read the original information.

An encryption method that uses the same secret key for both encryption and decryption, offering speed but requiring secure key distribution.

Encryption using a suitable public-key scheme and a corresponding private key for decryption. Authenticated key distribution and the specified padding/protocol remain essential; signatures and key agreement are distinct constructions.

A digest construction mapping input to an output under its defined algorithm, with cryptographic versions designed for appropriate resistance properties. Password storage needs a suitable salted password-hashing/KDF scheme rather than a bare fast digest.

An electronic signature that uses asymmetric cryptography to verify the authenticity and integrity of a message or document.

The policies and mechanisms that determine who can access which systems, data, or facilities and under what conditions.