Data Analytics and Visualization - Lecture Notes

A set of vocabulary-style flashcards covering key concepts from the lecture notes on data, analytics, quality, databases, time series, visualization, and SQL.

Data

Raw facts and figures used by computers to support decision making (input before processing).

Information

Processed data that informs decisions and actions.

Big Data

Extremely large and diverse data sets that require special methods to analyze.

6 V of Big Data

The six characteristics: Volume, Variety, Velocity, Veracity, Value, and Variability.

Volume

The amount of data available.

Variety

Different data types, including structured, semi-structured and unstructured data.

Velocity

The speed at which data is generated and processed.

Veracity

The degree of trustworthiness or reliability of data.

Value

Business value derived from data.

Variability

The ways data can be used or vary in interpretation.

Data analytics

The science of analyzing raw data and turning it into useful information.

Data analysis

A subset of data analytics focused on evaluating data to gain insights.

Descriptive analytics

Analyses that describe what happened in the data.

Diagnostic analytics

Analyses that explain why something happened.

Predictive analytics

Analyses that forecast what will happen in the future.

Prescriptive analytics

Analyses that suggest actions to make something happen.

Data culture

An organizational approach where decisions are data-informed, data-driven, and data-inspired.

Data-informed

Decisions guided by data insights.

Data-driven

Decisions primarily guided by data evidence.

Data-inspired

Decisions influenced by data but not solely determined by it.

Cognitive bias

Systematic errors in thinking that data helps reduce or counteract.

Competitive advantage

Gaining an edge over competitors through effective use of data.

Data literacy gap

96% of executives discount data they don't understand.

Problem Statement

Steps to frame business problems for data goals: understand situation, define goals, translate to data goals, frame problem, build metrics.

Data Wrangling

Cleaning, structuring, and enriching data for analysis.

Data quality

How good data is for use, including dimensions like accuracy, completeness, validity, consistency, integrity, and uniqueness.

Uniqueness

No duplicates; each value is distinct where required.

Accuracy

Data correctly reflects the real world or source.

Completeness

All required data is present.

Validity

Data conforms to defined formats and business rules.

Consistency

Data remains uniform across datasets and systems.

Integrity

Data relationships and structure are maintained accurately.

Data discovery

Understanding what the data says and implies.

Data structuring

Transforming data into a standard tabular format with variables (columns) and observations (rows).

Dataset

A collection of values organized for analysis.

Variable

A column representing a measured attribute.

Observation

A row representing measurements on a unit across attributes.

Relational database

A database organized into related tables with predefined relationships.

Database schema

Blueprint defining how data is organized and related in a relational database.

Time series data

Data points indexed in time order (time-stamped data).

Trend

General direction in data over time.

Cyclic variations

Regular cycles in data beyond seasonal patterns.

Seasonal variations

Regular patterns repeating within a fixed period (e.g., yearly).

Random movements

Unpredictable, irregular fluctuations in data.

Excel data types

Common data types in Excel including text, numbers, dates, times, boolean, and errors.

Text

Alphanumeric strings stored in cells.

Number

Numeric values (integers or decimals).

Date

Calendar date values.

Time

Time values (clock time).

Boolean

TRUE (1) or FALSE (0) values.

Error

Spreadsheet error values like #DIV/0, #N/A, etc.

SQL data types

Common SQL data types such as TEXT, CHAR, VARCHAR, NUMBER, FLOAT, INTEGER, BOOLEAN, DATE, TIME, DATETIME, TIMESTAMP, INTERVAL, NULL.

TEXT

Text data type in SQL for large strings.

CHAR

Fixed-length character data type in SQL.

VARCHAR

Variable-length character data type in SQL.

NUMBER

Numeric data type in some SQL dialects.

FLOAT

Floating-point numeric data type.

INTEGER

Whole-number numeric data type.

BOOLEAN

TRUE (1) or FALSE (0) data type.

DATE

Date value without time.

TIME

Time value without date.

DATETIME

Date and time value.

TIMESTAMP

Date and time value with possible timezone awareness.

INTERVAL

A duration of time in SQL.

NULL

A special value representing missing or unknown data.

Data Visualization

The representation of data through graphics like charts and infographics to communicate insights.

McCandless method

A framework by David McCandless for good data visualization focusing on data, story, goal, and visual form.