DATA SCIENCE (AD-404) Unit 5 Notes - RGPV

How unit 5 is examined

This unit covers what Business Intelligence (BI) is, its types and tools, and the ethical issues of data science; the marks sit in BI definition with types, BI tools, ethics and privacy, and the role of the data scientist.

Introduction

Medium weight

Definition. Business Intelligence is the set of processes, technologies and tools that collect, integrate, analyse and present business data as reports and dashboards so that managers can make better, faster decisions.

Diagram. <figure class="ds-fig" style="margin:1.4rem 0;overflow-x:auto"><svg xmlns="http://www.w3.org/2000/svg" id="dsfig-u5-01" viewBox="0 0 568.6 252" width="568.6" height="252" role="img" aria-label="BI architecture. Src = data sources, ETL = extract-transform-load, DW = data warehouse, OLAP = multidimensional analysis, Rep = reporting, Dash = dashboards"><style>#dsfig-u5-01 .e{stroke:#454C5A;stroke-width:1.4;fill:none}#dsfig-u5-01 .e.hi{stroke:#2340B8;stroke-width:2.6}#dsfig-u5-01 .n{fill:#FFFFFF;stroke:#16181D;stroke-width:1.4}#dsfig-u5-01 .n.hi{fill:#E3E9FC;stroke:#2340B8;stroke-width:2.2}#dsfig-u5-01 .n.rb-b{fill:#16181D;stroke:#16181D}#dsfig-u5-01 .n.rb-r{fill:#BD3227;stroke:#BD3227}#dsfig-u5-01 text{font-family:"JetBrains Mono",ui-monospace,Menlo,Consolas,monospace;font-size:13px}#dsfig-u5-01 .t{fill:#16181D;font-weight:500}#dsfig-u5-01 .t.inv{fill:#FFFFFF;font-weight:700}#dsfig-u5-01 .kd{stroke:#16181D;stroke-width:1.2}#dsfig-u5-01 .dot{fill:#16181D}#dsfig-u5-01 .ann{fill:#2340B8;font-size:11px;font-weight:700}#dsfig-u5-01 .lbl{fill:#6F7787;font-family:system-ui,-apple-system,sans-serif;font-size:12px;font-weight:700}#dsfig-u5-01 .ptr{fill:#2340B8;font-size:12px;font-weight:700}#dsfig-u5-01 .ah{fill:#454C5A}#dsfig-u5-01 .ah.hi{fill:#2340B8}#dsfig-u5-01 .wl rect{fill:#FFFFFF;stroke:#DCE0E7}#dsfig-u5-01 .wl .t{font-size:12px;font-weight:700}#dsfig-u5-01 .wl.hi rect{fill:#2340B8;stroke:#2340B8}#dsfig-u5-01 .wl.hi .t{fill:#FFFFFF}html.dark #dsfig-u5-01 .e{stroke:#B1B7C3}html.dark #dsfig-u5-01 .e.hi{stroke:#8FA3FF}html.dark #dsfig-u5-01 .n{fill:#161920;stroke:#E6E8ED}html.dark #dsfig-u5-01 .n.hi{fill:#1E2748;stroke:#8FA3FF}html.dark #dsfig-u5-01 .n.rb-b{fill:#E6E8ED;stroke:#E6E8ED}html.dark #dsfig-u5-01 .n.rb-r{fill:#FF7E71;stroke:#FF7E71}html.dark #dsfig-u5-01 .t{fill:#E6E8ED}html.dark #dsfig-u5-01 .t.inv{fill:#0F1115}html.dark #dsfig-u5-01 .kd{stroke:#E6E8ED}html.dark #dsfig-u5-01 .dot{fill:#E6E8ED}html.dark #dsfig-u5-01 .ann{fill:#8FA3FF}html.dark #dsfig-u5-01 .lbl{fill:#858D9C}html.dark #dsfig-u5-01 .ptr{fill:#8FA3FF}html.dark #dsfig-u5-01 .ah{fill:#B1B7C3}html.dark #dsfig-u5-01 .ah.hi{fill:#8FA3FF}html.dark #dsfig-u5-01 .wl rect{fill:#161920;stroke:#2A2E37}html.dark #dsfig-u5-01 .wl.hi rect{fill:#8FA3FF;stroke:#8FA3FF}html.dark #dsfig-u5-01 .wl.hi .t{fill:#0F1115}</style><defs><marker id="ah3" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="7" markerHeight="7" orient="auto-start-reverse"><path class="ah" d="M0,1 L9,5 L0,9 z"/></marker><marker id="ahh3" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="7" markerHeight="7" orient="auto-start-reverse"><path class="ah hi" d="M0,1 L9,5 L0,9 z"/></marker></defs><path class="e" d="M59,126 L139.4,126" marker-end="url(#ah3)"/><path class="e" d="M179.4,126 L259.8,126" marker-end="url(#ah3)"/><path class="e" d="M296.3,115 L378.4,56.3" marker-end="url(#ah3)"/><path class="e" d="M296.3,137 L384.1,199.8" marker-end="url(#ah3)"/><path class="e" d="M422.4,55.1 L498.8,109.7" marker-end="url(#ah3)"/><path class="e" d="M416.7,201 L498.8,142.3" marker-end="url(#ah3)"/><circle class="n" cx="40" cy="126" r="18"/><text class="t" x="40" y="126" dy=".35em" text-anchor="middle">Src</text><circle class="n" cx="160.4" cy="126" r="18"/><text class="t" x="160.4" y="126" dy=".35em" text-anchor="middle">ETL</text><circle class="n" cx="280.8" cy="126" r="18"/><text class="t" x="280.8" y="126" dy=".35em" text-anchor="middle">DW</text><rect class="n" x="376.2" y="25" width="50" height="30" rx="15"/><text class="t" x="401.2" y="40" dy=".35em" text-anchor="middle">OLAP</text><circle class="n" cx="401.2" cy="212" r="18"/><text class="t" x="401.2" y="212" dy=".35em" text-anchor="middle">Rep</text><rect class="n" x="496.6" y="111" width="50" height="30" rx="15"/><text class="t" x="521.6" y="126" dy=".35em" text-anchor="middle">Dash</text></svg><figcaption style="font-size:.82em;opacity:.72;margin-top:.45rem">BI architecture. Src = data sources, ETL = extract-transform-load, DW = data warehouse, OLAP = multidimensional analysis, Rep = reporting, Dash = dashboards</figcaption></figure>

Key points.

Data sources such as ERP, CRM, sales databases, spreadsheets and web logs feed the BI system.
ETL extracts data from the sources, cleans and transforms it, and loads it into a central data warehouse.
The data warehouse stores integrated, historical, subject-oriented data for analysis.
OLAP lets users slice, dice, drill down and roll up the data across dimensions such as time, region and product.
Reporting, dashboards and data mining present the results as charts, KPIs and patterns to decision makers.
Example: a retail chain uses BI to see which products sell in which city each month and reorders stock accordingly.

Issues and challenges.

Poor data quality (duplicates, missing values, inconsistent formats) gives wrong insights.
Integrating many heterogeneous sources into one warehouse is difficult.
Scalability suffers as data volume and number of users grow.
Cost of tools, infrastructure and licences is high.
Data security and privacy must be protected.
Skilled people and user adoption are scarce, and poor BI leads to wrong decisions.

Answer frame. Open with the definition; draw the architecture diagram; develop components in flow order, then examples; for the issues question list challenges 1-6 and close with their effect on decision making.

Asked: [7 marks] (Dec 2024) Explain the major issues and challenges in Business Intelligence. Asked: [7 marks] (Jun 2026) What is Business Intelligence also discuss its types and component.

Types of Business Intelligence

Definition. BI types are the kinds of analysis BI performs, by purpose: descriptive, diagnostic, predictive and prescriptive, and by decision level: strategic, tactical and operational.

Key points.

Descriptive BI reports what has happened using historical data, for example last quarter's sales dashboard.
Diagnostic BI explains why it happened through drill-down and root-cause analysis.
Predictive BI forecasts what is likely to happen using statistics and machine learning, for example demand forecasting.
Prescriptive BI recommends the best action, for example the optimal price or stock level.
Strategic BI supports long-term, top-management decisions such as entering a new market.
Tactical BI supports middle-management decisions such as monthly budgets, and operational BI supports daily decisions such as order tracking.

Strategic decision support (logic).

Step 1: Collect data from internal and external sources.
Step 2: ETL into the data warehouse.
Step 3: Analyse with OLAP, KPIs and predictive models.
Step 4: Present trends on dashboards to top management.
Step 5: Managers choose a strategy, then monitor results with the same KPIs.

Example: sales trends show region East is growing, so management opens new stores there.

Answer frame. Open with the BI definition; give components briefly; explain the types in the order descriptive, diagnostic, predictive, prescriptive, then levels; for Jun 2025 add the workflow with the example; close with business applications.

Asked: [9 marks] (Jun 2024) What is Business Intelligence? Explain the types of Business Intelligence. Asked: [7 marks] (Jun 2025) Define Business Intelligence (BI). Explain its types and write the logic of how BI supports strategic decision-making.

Modern Business Intelligence Tools

Definition. BI tools are software applications that connect to data, model it and turn it into interactive reports and dashboards; modern tools such as Power BI, Tableau and Qlik are self-service, visual and cloud-ready.

Key points.

Power BI (Microsoft) integrates with Excel and Azure and is cheap and easy for business users.
Tableau offers the most powerful drag-and-drop visualisation for large datasets.
Qlik Sense uses an associative engine that lets users explore links across data freely.
Others: Looker (Google), SAP BusinessObjects, IBM Cognos and Metabase.
Common features are dashboards, ad-hoc queries, data connectors, sharing and mobile access.
Use cases: sales tracking, financial reporting, customer segmentation and supply-chain monitoring.

Basis	Power BI	Tableau
Usability	Familiar Excel-like interface, easy for beginners	Steeper learning curve, drag-and-drop for analysts
Visualisation	Good standard visuals plus custom visuals	Richer, highly customisable visuals
Data handling	Handles moderate data well; DAX for modelling	Handles very large data with live or extract connections
Connectors	Excel, SQL, Azure, cloud services	Wide range of databases and files
Cost	Lower, Pro licence per user	Higher licence cost
Best for	Microsoft-based organisations	Deep visual exploration

Answer frame. For the overview: define BI and its need, sketch the architecture, list the tools with features, close with use cases. For the comparison: choose Power BI and Tableau, use the table on usability, features and data handling, and close with a recommendation.

Asked: [9 marks] (Jun 2023, Jun 2026) Give a brief overview on Business Intelligence and Business Intelligence Tools. Explain Modern Business Intelligence tools in detail. Asked: [7 marks] (Jun 2025) Compare any two modern Business Intelligence tools based on usability, key features, and data handling capabilities.

Modern Business Intelligence

Definition. Modern BI is self-service, cloud-based analytics that lets business users explore data and get real-time insights without depending on IT.

Key points.

Self-service means non-technical users build their own reports and dashboards by drag and drop.
Cloud deployment removes hardware cost and scales on demand.
Real-time data and streaming replace the periodic batch reports of traditional BI.
It uses AI, machine learning and natural-language queries for automated insights, with tools such as Power BI and Tableau.
Benefit over traditional BI: faster, cheaper, more collaborative and accessible on mobile.

Asked: [5 marks] (Dec 2024) Explain the following in detail: Modern Business Intelligence

Data Science and Ethical Issues: Unfair discrimination

Definition. Unfair discrimination in data science is when a model or dataset systematically treats people worse because of attributes such as gender, race, age or religion.

Key points.

Biased or unrepresentative training data teaches the model the same skew.
Algorithm design and proxy variables, such as postcode standing in for race, can discriminate indirectly.
Examples: a hiring model rejecting women applicants, or a loan model denying a minority area.
Mitigation: use balanced data, audit fairness metrics, remove proxies and keep human review.

Asked: [5 marks] (Jun 2024) What is unfair discrimination? Explain.

Reinforcing human biases

Definition. Reinforcing human bias means models learn prejudices present in historical human decisions and then repeat and amplify them at scale.

Key points.

Historical data records past human prejudice, so a model trained on it treats that bias as truth.
Predictions feed back into decisions, creating a loop, for example predictive policing sends more police to an area, which records more crime there.
Examples: a recruitment tool trained on male-dominated hires downgrades women, and face recognition errs on darker skin.
Mitigation: diverse data and teams, bias testing, fairness constraints and regular audits.

Asked: [5 marks] (Dec 2024) Explain the following in detail: Reinforcing human biases

Lack of transparency

Definition. Lack of transparency is the black-box problem: a complex model gives predictions without a clear explanation of how it reached them.

Key points.

Users and regulators cannot check or challenge a decision, such as a loan rejection, that has no reason.
It hides bias and errors and reduces trust and accountability.
Explainable AI (XAI) methods such as LIME and SHAP, plus simple interpretable models, documentation and audit trails, address it.

Discussions on privacy, security, ethics

Definition. Ethics in data science is the set of moral principles that govern how personal data is collected, analysed and used, so that people are not harmed.

Ethical issues.

Privacy: personal data such as location or purchases must not be collected or exposed without protection.
Informed consent: people must know and agree how their data will be used.
Bias and discrimination: models can treat groups unfairly.
Transparency: decisions should be explainable.
Accountability and ownership: someone must answer for harm, and it must be clear who owns the data.
Examples of violations: Cambridge Analytica misusing Facebook data, and selling customer data without consent.
Best practices: data minimisation, consent, anonymisation, fairness audits, ethics review and laws such as GDPR.

Security issues.

Threats: data breaches, unauthorised access, insider misuse, data leakage and ransomware.
Measures: encryption, access control and authentication, anonymisation and masking, audit logs, backups and secure cloud practice.
Example: hackers steal a hospital's patient records, so encryption and role-based access reduce the damage.

Answer frame. Open by defining ethics; list issues 1-5 with a one-line example each, then best practices; for security, define the need, give threats, then measures, and close with an example.

Asked: [7 marks] (Jun 2023, Jun 2024) What are the ethical issues in Business Intelligence? Discuss. Explain the following in detail: Ethical Issues in Data Science Asked: [5 marks] (Jun 2024) Security issues in Data Science

Role of Next-generation data scientists

Definition. A data scientist extracts insight from data using statistics, programming and domain knowledge and communicates it to support business decisions; a next-generation data scientist also works with AI, big data and automation and acts ethically.

Key points.

Responsibilities: collect, clean and prepare data.
Explore data and build, test and deploy machine learning models.
Communicate findings through visualisation and storytelling to non-technical managers.
Skills: statistics, Python or R, SQL, machine learning, big-data tools and visualisation.
Next-generation scientists use AutoML, deep learning, cloud and MLOps, and work on AI and real-time big data.
They collaborate with the business, understand domain problems, and ensure privacy, fairness and explainability.
Business value: better decisions, cost savings, new products and competitive advantage.

Answer frame. Open with the definition; develop responsibilities, skills, next-generation trends, ethics and collaboration in that order; close with business value.

Asked: [7 marks] (Jun 2023, Dec 2024, Jun 2026) Write a short note on the role of Data Scientist. Explain the role of next-generation data scientists in detail.

Last-minute revision

BI turns raw data into reports and dashboards for decisions.
BI flow: sources, ETL, data warehouse, OLAP, reporting, dashboards.
Purpose types: descriptive, diagnostic, predictive, prescriptive.
Level types: strategic, tactical, operational.
Main tools: Power BI, Tableau, Qlik.
Power BI is easy and cheap; Tableau is stronger on visualisation.
Modern BI: self-service, cloud, real-time, AI.
Bias comes from data and algorithms and is amplified by feedback loops.
Black box is solved by XAI (LIME, SHAP).
Security measures: encryption, access control, anonymisation.
Ethics: privacy, consent, bias, transparency, accountability.

Memory hooks

Types: "Do Data Predict Prescriptions" gives descriptive, diagnostic, predictive, prescriptive.
Flow: "Sources Extract, Warehouse Queries, Reports Display" (Src, ETL, DW, OLAP, Report, Dashboard).
Ethics: "PCBTA" for privacy, consent, bias, transparency, accountability.
Security: "Encrypt, Control, Anonymise".

Coverage checklist

Introduction: BI definition, architecture, challenges; covers Dec 2024 and Jun 2026 questions.
Types of Business Intelligence: covers Jun 2024 and Jun 2025 questions.
Modern Business Intelligence Tools: overview and Power BI vs Tableau comparison.
Modern Business Intelligence: covers Dec 2024.
Data Science and Ethical Issues- Unfair discrimination: covers Jun 2024.
Reinforcing human biases: covers Dec 2024.
Lack of transparency: definition and XAI.
Discussions on privacy, security, ethics: ethical issues and security issues.
Role of Next-generation data scientists: covers Jun 2023, Dec 2024, Jun 2026.