#1 (VOCAB) Chapter 2: Data Management

Objectives: Big data to smart data: Database management Systems, relational databases, structured query languages: Key elements of enterprise data architecture: Dimensions of data quality and the importance of marketing analytics: Preparing data prior to analyzing:

big data

Big data is the term typically used to describe massive amounts of data. It is a relative term that requires special tools to manage and analyze.

volume

Volume refers to large amounts of data per time unit.

velocity

Velocity refers to the rapid pace in which data is being collected or is available.

veracity

Veracity refers to the messy data that could have missing values, inconsistencies in the unit of measurement, erroneous information, or lack of reliability.

variety (data)

Variety refers to the variety of the dataset. Holistic and diverse data provides a better understanding of customers and market situations.

value

Value means the data is useful for making accurate business decisions.

smart data

Smart data represents data that is valuable and can be effectively used.

database

A database contains collected data from company operations.

relational database

A relational database is a type of database management system (DBMS). It is a collection of interrelated data items organized using software programs to manipulate and access data.

Structured Query Language (SQL)

Structured querying language (SQL) is the language developed by IBM and is used to access and update data stored in the database.

streaming data

Streaming data is the continuous transfer of data from numerous sources in different formats.

Hadoop

Hadoop is an open-source software that helps distributed computers solve problems of big data computation. Hadoop divides the big data processing over multiple computers, allowing it to handle massive amounts of data simultaneously at a reduced cost.

data warehouse

A data warehouse contains historical data from various databases throughout a company and provides a structured environment for high-speed querying. A data warehouse consists of data from different functional areas of the firm such as customer data, accounting data, and human resources data.

data mart

A data mart is a subset of the data warehouse that provides a specific value to a group of users. For example, a marketing data mart would be limited to data on customers, sales, products, and similar marketing metrics. There are two types of data marts: dependent and independent.

Data lake

A data lake (often included in Hadoop systems) is a storage repository that holds a large amount of data in its native format. It is typically used to investigate data patterns and to archive data for future use.

data management

Data management as a process is the lifecycle management of data from acquisition to disposal.

data quality

Data quality refers to attributes such as timeliness, completeness, accuracy, consistency, and format of the dataset.

feature

Feature is another name for variable.

overfitting

Overfitting occurs from an overly complex model where the results are limited to the data being used and are not generalizable—which means future relationships cannot be inferred, and results will be inconsistent when using other data.

sample size

A sample size is a portion of data from the entire population.

power calculation

A power calculation helps determine that outcomes will be estimated from a sample with a sufficient level of precision.

unit of analysis

A unit of analysis describes the what, when, and who of the analysis.

outlier

A value that is at a considerable distance from any of the other data clusters.

Cluster analysis

Cluster analysis is an analytical method of segmenting a market based on identifying shared characteristics of groups of individuals.

aggregation

Aggregation refers to the summing of data according to the unit of analysis desired. For example, summing all weekly data into quarterly data.

normalization

Normalization helps bring all variables into the same scale. To normalize a variable, we scale it by subtracting the variable from the mean and then dividing it by the standard deviation.

categorical variables

A categorical variable is when the data represents one of a limited number of categories. Geographic location (e.g., Northeast, Southeast, Northwest, Southwest, Midwest) is an example of a categorical variable.