______ is process of extracting previously non known valid and actionable…

2012

______ is process of extracting previously non known valid and actionable information from large data to make crucial business and strategic decisions.

Answer: C. Data MiningConcept: An enterprise data stack separates into layers with different jobs — a layer that stores data, a layer that describes data, a layer that governs…

  1. A.

    Data Management

  2. B.

    Data base

  3. C.

    Data Mining

  4. D.

    Meta Data

Attempted by 13 students.

Show answer & explanation

Correct answer: C

Concept: An enterprise data stack separates into layers with different jobs — a layer that stores data, a layer that describes data, a layer that governs data, and a layer that discovers knowledge from data. Only the discovery layer yields information that is not already recorded anywhere: it runs statistical and machine-learning algorithms over large historical data sets to derive validated patterns, and those derived patterns are what support business and strategic decisions.

Application: Three markers in the sentence fix which layer the blank names.

  1. "previously non known" — the output has to be new knowledge rather than a record that is already held, so the blank names an activity that derives something, not one that keeps or documents what exists.

  2. "valid and actionable" — the output is a statistically validated pattern that can be acted upon, which is the product of running an analytic algorithm over the data.

  3. "from large data … crucial business and strategic decisions" — the input is a large historical data store and the consumer is decision support, which is exactly the role of the discovery layer.

Together these markers describe the discovery layer of the data stack, whose name is Data Mining.

Cross-check: placing all four terms on the same map shows that only one of them is a discovery activity.

Term

Layer of the data stack

What it produces

Data Management

Governance and lifecycle

Policies, standards and controls over data assets

Database

Storage

Reliable persistence, retrieval and update of records

Data Mining

Discovery and analysis

Validated patterns and models derived from stored data

Metadata

Description

Documentation of schema, format, ownership and lineage

Data mining is the analysis step of the wider KDD (Knowledge Discovery in Databases) process, which also includes selection, cleaning, transformation and interpretation of the results.

Explore the full course: Mppsc Assistant Professor Computer Science Paper 2

Loading lesson…