An Overview of Business Intelligence, Analytics, and Data Science (copy)

Comprehensive vocabulary flashcards covering the evolution of computerized decision support, business intelligence frameworks, the three types of analytics, and industry applications from Chapter 1 of IDM 2020.

Sports Analytics Industry Growth

The expansion of analytics in the sports sector, which generates approximately $500\,\text{billion}$ in annual revenues, with industry analytics expenditure growing from $125\,\text{million}$ in 2018 to $885\,\text{million}$ in 2020.

Six Areas of Sports Analytics

A framework categorizing sports analytics into six distinct domains: Administrator Analytics, Fan/Donor Analytics, Roster/Recruiting Analytics, Team Tactics Analytics, Health/Safety Analytics, and League/Conference Analytics.

Moneyball

A 2003 book by Michael Lewis (and follow-on 2011 movie) about the Oakland A's that popularized the adoption of data analytics across professional sports franchises.

Fan Analytics

A business office analytics application focused on predicting season ticket renewals through fan satisfaction survey scores, customer sentiment comments, seating tiers, and team performance.

Dynamic Pricing Variables (Major League Baseball)

Key factors used to adjust ticket prices dynamically, including Seat Location, Team Performance (past 10 games, opponent playoff status), Time-Related Variables (start time, season phase, days before game), Division Rivalry, and Individual Player Reputations (starting pitcher, ERA, opponent All-Stars).

Heat Map Zone Analysis

A spatial analytical tool used by football coaches to evaluate passing play efficiency, completion percentages, and explosive pass plays across distinct field zones relative to the line of scrimmage.

Football Opponent Analysis Factors

Quantitative factors analyzed to anticipate an opponent's play-calling tendencies, including Personnel Formations (e.g., '11', 'GULA'), Down and Distance, and Field Position (e.g., Danger Zone, Midfield, Red Zone).

Cricket Player Analytics

Coaching and roster analytics that evaluate a batter's scoring rate, shot placement, and vulnerability across delivery lengths (Yorker, Full Toss, Good Length, Short Length) and bowling types (Googly, LegBreak, OffBreak).

Health and Safety Analytics in Sports

Predictive modeling and monitoring applications aimed at preventing athlete overtraining, managing physical fatigue, mitigating injury risks, and conducting concussion research.

Roster and Recruitment Analytics

The evaluation of high school and prospective recruits using cognitive and specialized physical metrics (such as head rotation speed and reaction times to stimuli) in addition to traditional performance statistics.

Administrative and Scheduling Analytics in Sports

The application of optimization algorithms and genetic algorithms to automate conference tournament seeding, regional assignments, travel schedules, and venue operations.

Simon's Decision-Making Process

A classical four-phase conceptual model of decision making comprising Intelligence (problem identification), Design (modeling and formulating alternatives), Choice (selecting the best alternative), and Implementation.

Intelligence Phase

The first phase of Simon's decision-making process involving environmental scanning, data collection, identifying organizational problems or opportunities, and establishing problem ownership.

Problem Classification

The conceptualization of an identified problem in order to place it into an established, recognizable category that suggests appropriate solution methods.

Problem Decomposition

The practice of breaking down a large, complex, or unstructured problem into smaller, more manageable subproblems to facilitate analysis and modeling.

Problem Ownership

The formal assignment of organizational responsibility, authority, and commitment to a person or group to address and solve an identified problem.

Design Phase

The second phase of Simon's decision-making process where decision makers formulate a mathematical or conceptual model, establish criteria for choice, search for alternatives, and predict potential outcomes.

Choice Phase

The third phase of Simon's decision-making process involving generating solutions to the model, conducting sensitivity analysis, and selecting the best or most satisfactory course of action.

Implementation Phase

The final stage of Simon's decision-making framework where the selected alternative is put into operational practice, followed by verification and testing against real-world performance.

Management Information Systems (MIS)

Early computer-based information systems from the 1970s designed to generate periodic, structured, two-dimensional reports summarizing historical organizational transactions for managers.

Decision Support Systems (DSS)

Interactive computer-based systems developed in the 1970s that combine databases and analytical decision models to assist managers in solving semi-structured and unstructured problems.

Operations Research (OR)

A scientific and mathematical discipline utilizing mathematical optimization, simulation, and heuristics to determine optimal or satisfactory solutions with limited computational data.

Expert Systems (ES)

Rule-based artificial intelligence systems developed in the 1980s designed to capture, store, and apply the knowledge and heuristics of human specialists to solve complex domain problems.

Enterprise Resource Planning (ERP)

Integrated transactional enterprise software platforms introduced to consolidate data across corporate functions (e.g., finance, supply chain, HR) into a unified, consistent database.

Executive Information Systems (EIS)

Specialized executive-level reporting systems developed in the 1980s featuring visual dashboards, key performance indicators (KPIs), ad hoc reporting, drill-down capabilities, and trend analysis.

Drill-Down

An interactive data exploration capability within executive dashboards and BI tools that allows users to click on high-level summary metrics to view increasingly granular, underlying detailed data.

Evolution of Decision Support, BI, Analytics, and AI

The multi-decade progression of decision technologies: Decision Support Systems (1970s), Enterprise/Executive IS (1980s), Business Intelligence (1990s), Analytics (2000s), Big Data (2010s), to AI and Automation (2020s).

Business Intelligence (BI)

An umbrella term coined by Gartner in the mid-1990s that combines architectures, tools, databases, and analytical methodologies to enable interactive access to data; today, it represents descriptive analytics.

High-Level Architecture of BI

A three-tier structure connecting data sources to a Data Warehouse Environment (built by technical staff), a Business Analytics Environment (accessed by business users), and Performance/Strategy (used by managers/executives).

Online Transaction Processing (OLTP)

Operational database systems (such as ERP, CRM, and SCM) optimized for fast, real-time data capture, updating, and routine daily business transactions.

Online Analytical Processing (OLAP)

Multidimensional database analysis technologies and data warehouses designed to support complex querying, data exploration, and executive decision making rather than transaction recording.

Extract, Transform, Load (ETL)

The data integration process of extracting raw data from multiple operational source systems, cleaning and standardizing it, and loading it into a centralized data warehouse or data mart.

Data Warehouse (DW)

A centralized, integrated, subject-oriented repository of historical organizational data structured specifically to support multidimensional querying, reporting, and business analytics.

Data Mart

A focused, departmental or subject-specific subset of a data warehouse designed to address the reporting and analytical needs of a single business unit or functional group.

Metadata

Descriptive 'data about data' that provides technical definitions, lineage, structures, and business context for information assets stored in a database or data warehouse.

BI Competency Center (BICC)

A dedicated organizational entity that aligns BI practices with corporate strategy, bridges business units with IT, and serves as a repository for disseminating analytical best practices.

Analytics (INFORMS Definition)

As defined by the Institute for Operations Research and the Management Sciences (INFORMS), the combination of computer technology, management science techniques, and statistics to solve real-world problems.

Three Types of Analytics

The standard categorization of business analytics into Descriptive Analytics ('What happened?'), Predictive Analytics ('What will happen?'), and Prescriptive Analytics ('What should I do?').

Descriptive Analytics

The branch of analytics focused on retrospective analysis of historical data to answer 'What happened?' and 'What is happening?', enabled by data visualization, dashboards, scorecards, OLAP, and descriptive statistics.

Predictive Analytics

The branch of analytics aimed at determining what is likely to happen in the future by applying statistical modeling, data mining, text/web mining, and time-series forecasting to historical data.

Prescriptive Analytics

The branch of analytics (also known as decision or normative analytics) that evaluates alternative courses of action to recommend the best possible decision, enabled by optimization, simulation, and decision modeling.

PANDA Platform

A cloud-based analytics architecture developed for the Purina Research Facility on Microsoft Azure Cloud, integrating data acquisition software, transactional layers (Blob storage/SQL), analytical data warehouses, and Power BI dashboards.

Machine Learning Ensemble (Purina Cattle Trial)

A predictive modeling architecture that combines and averages predictions from Ridge Linear Regression, Gradient Boosting, Random Forests, and Support Vector Machines to forecast final animal weight after $63\,\text{days}$ on trial.

Retail Value Chain Analytics

The comprehensive application of analytics across all critical retail touchpoints, encompassing Vendors, Planning, Merchandising, Buying, Warehouse & Logistics, Multichannel Operations, and Customers.

Inventory Optimization

A retail analytics application that forecasts the consumption of fast-moving products to prevent stock-outs and pairs slow-moving items with high-demand goods to accelerate inventory turnover.

Price Elasticity Analytics

The mathematical modeling of pricing sensitivity to optimize product markdown schedules, determine acceptable discount boundaries, and protect net gross margin dollars.

Market Basket Analysis

An affinity analysis technique in retail that identifies hidden associations and purchase correlations among products to design bundle offers and optimize cross-selling and up-selling strategies.

Gaming Analytics Platform

An AI-powered casino enterprise intelligence system (such as Gaming Analytics.ai used at Gulfstream Park) utilizing knowledge graphs and NLP to optimize slot machine placement, calculate slot ROI, and detect problem gambling.

COVID-19 Time Series Analytics

Public health epidemiology modeling that evaluates infection rates and per capita case volumes over time, such as tracking differences in infection curves between mask-mandated and non-mask-mandated cities in Oklahoma.

Data Analyst vs. Data Scientist

A comparison of roles: a Data Analyst primarily examines historical structured data to explain past business trends using SQL, Excel, and Power BI; a Data Scientist develops machine learning models using Python, R, and Spark on structured and unstructured data to predict and automate outcomes.