BAT 401 Prelims Reviewer - Fundamentals of Business Analytics

Vocabulary practice flashcards generated from the BAT 401 Prelims Reviewer notes.

Business Analytics (BA)

Using data to help a business make better decisions (DATA → UNDERSTAND → DECIDE).

Data

Raw information that is collected, transformed, or organized.

Analysis

Examining something carefully to understand it and draw conclusions.

Analytics

Using data, tools, statistical methods, and technology to generate useful insights.

Data Analytics (DA)

Finds patterns, trends, relationships, and insights (KEYWORD: patterns).

Business Intelligence (BI)

Uses reports, dashboards, and KPIs to understand business information (KEYWORD: dashboards/reports).

Machine Learning (ML)

Uses algorithms that learn from data for predictions or automation (KEYWORD: learn + predict).

Descriptive Analytics

Looks at past/historical data to determine 'WHAT HAPPENED?' (KEYWORD: PAST).

Predictive Analytics

Forecasts future outcomes to determine 'WHAT IS LIKELY TO HAPPEN?' (KEYWORD: FUTURE).

Prescriptive Analytics

Recommends actions to answer 'WHAT SHOULD WE DO?' (KEYWORD: ACTION).

Prehistoric Era (BA Evolution)

Era focused on survival data collection using tally sticks and cave paintings (KEYWORD: survival).

Industrial Age (BA Evolution)

1800s era characterized by factories, mass production, accounting, ledgers, and early statistics (KEYWORD: factories).

Information Age (BA Evolution)

Mid-20th century era marked by the growth of information and computing.

Internet Age (BA Evolution)

21st century era characterized by a huge growth of digital data, social media, mobile apps, Big Data, AI, and cloud (KEYWORD: digital data).

Big Data

Very large amounts of data, supported by providers like Google, Meta, Microsoft, Amazon/AWS, IBM, and government agencies.

Internal Data

Data coming from INSIDE the organization, such as sales, employees, inventory, HR, CRM, POS, and financial reports.

External Data

Data coming from OUTSIDE the organization, such as government statistics, market research, social media, news, suppliers, weather, World Bank, and Kaggle.

Primary Data

Data collected first-hand for a specific purpose.

Secondary Data

Data already collected by someone else and reused.

Nominal Level of Measurement

NAME/CATEGORY data with no order (e.g., blood type, nationality, hair color, course, car brand).

Ordinal Level of Measurement

ORDER/RANK data that has order, but differences are not exact (e.g., class rank, satisfaction level, performance grade).

Interval Level of Measurement

Data with equal intervals but NO TRUE ZERO (e.g., Celsius/Fahrenheit, IQ, dates/years, time of day, GPA).

Ratio Level of Measurement

Data with equal intervals AND TRUE ZERO (e.g., age, weight, height, income, distance, number of children, sales, population).

Structured Data

Well-organized data, usually organized into rows and columns (e.g., spreadsheets, CSV, relational databases/SQL; KEYWORD: table).

Semi-structured Data

Data with some organization, but not a strict table format (e.g., JSON, XML, HTML; KEYWORD: some structure).

Unstructured Data

Data with no predefined/fixed structure (e.g., images, videos, audio, social posts, emails, PDFs, Word, PowerPoint, chats; KEYWORD: no fixed structure).

Data Cleaning

Fixing or removing incorrect, duplicate, or irrelevant data to reduce errors and improve reliability.

Missing Data

A data quality issue where a value is blank or unavailable.

Duplicate Data

A data quality issue where the same record appears more than once, either as exact or partial duplicates.

Inconsistent Data

A data quality issue involving different formats, typos, or logical errors (e.g., PH / Philippines / P.H. or negative age).

Outlier

A value very different from the others, which may be indicated by a Z-score above $3$ or below $-3$, or by using the $1.5 \times IQR$ rule.

Listwise Deletion

Deleting the WHOLE ROW when it contains missing data.

Pairwise Deletion

Leaving out missing values only for the SPECIFIC analysis where they are needed, keeping more data intact.

Mean Imputation

Filling a missing numeric value with the AVERAGE.

Median Imputation

Filling a missing value with the MIDDLE value; particularly useful for skewed data.

Mode Imputation

Filling a missing categorical value with the MOST COMMON value.

KNN Imputation

Using SIMILAR records/neighbors to estimate a missing value.

Regression Imputation

Using a model to PREDICT a missing value from other variables.