Microsoft Advanced Data Analysis with Generative AI Course

Lesson 1 • Generative AI Concepts for Analysts

Covers large language models, tokens, and prompt-response mechanics. Grounds analysts in how AI generates outputs from structured and unstructured data inputs.

Lesson 2 • Data Readiness and AI Compatibility

Examines how data structure, quality, and format affect AI performance. Students learn to assess and prepare datasets before applying generative AI techniques.

Lesson 3 • Ethical and Governance Foundations

Introduces bias, privacy, and accountability principles relevant to AI-assisted analysis. Establishes a responsible-use framework applied throughout the course.

Lesson 1 • Anatomy of an Effective Analytical Prompt

Breaks down the components of high-quality prompts: role, context, task, and format. Students apply a repeatable structure to analytical queries.

Lesson 2 • Prompting for Data Exploration

Teaches prompt patterns that surface trends, outliers, and distributions in datasets. Connects prompt design directly to exploratory data analysis goals.

Lesson 3 • Prompt Templates and Reusability

Covers building modular, parameterised prompt templates for recurring analytical tasks. Students create a personal prompt library for consistent, efficient AI use.

Lesson 4 • Chain-of-Thought and Multi-Step Prompting

Introduces reasoning-chain techniques that guide AI through complex analytical steps. Students decompose multi-part problems into sequenced prompt workflows.

Lesson 1 • Standardising and Normalising Data

Leverages AI to detect inconsistent formats, units, and naming conventions across datasets. Students apply AI-generated transformation scripts to enforce data standards.

Lesson 2 • Detecting and Handling Missing Data with AI

Uses AI prompts to identify missing value patterns and recommend imputation strategies. Connects automated detection to informed, context-aware data decisions.

Lesson 3 • Duplicate Detection and Record Deduplication

Applies AI to surface exact and fuzzy duplicate records across large datasets. Students implement AI-assisted deduplication workflows with human review checkpoints.

Lesson 1 • AI-Guided Statistical Summarisation

Applies AI to generate descriptive statistics and interpret their analytical significance. Bridges raw numeric output with narrative insight for business audiences.

Lesson 2 • Correlation and Relationship Discovery

Uses AI to identify and explain variable relationships within structured datasets. Students distinguish correlation from causation using AI-assisted reasoning prompts.

Lesson 3 • Automated EDA Report Generation

Generates structured exploratory reports using AI from raw dataset inputs. Students customise AI-produced EDA reports for specific stakeholder audiences.

Lesson 4 • Segmentation and Clustering with AI

Applies AI to propose and interpret data segmentation schemes from behavioural or demographic data. Students validate AI-suggested clusters against business logic.

Lesson 1 • Generating Visualisation Code with AI

Applies AI to write and refine chart code in Python, R, or spreadsheet formulas. Students iterate on AI-generated code to meet precise visual specifications.

Lesson 2 • Interactive Dashboard Design with AI

Applies AI to plan dashboard layouts, KPI hierarchies, and filter logic. Students produce dashboard specifications ready for implementation in BI tools.

Lesson 3 • Selecting the Right Chart with AI Guidance

Uses AI to recommend chart types based on data structure and analytical intent. Students apply selection logic to avoid misleading or ineffective visualisations.

Lesson 1 • Generating and Refining Model Code with AI

Applies AI to write, debug, and optimise machine learning code across common frameworks. Students validate AI-generated code against expected model behaviour.

Lesson 2 • Model Evaluation and Interpretation

Uses AI to explain model performance metrics and surface actionable insights from results. Students translate technical evaluation outputs into business-relevant conclusions.

Lesson 3 • Forecasting and Time-Series Analysis with AI

Applies AI to identify seasonality, trends, and forecast future values in time-series data. Students produce and validate AI-assisted forecasts for operational planning.

Lesson 4 • AI-Assisted Model Selection

Uses AI to evaluate and recommend modelling approaches based on data characteristics and business goals. Students apply structured criteria to justify model choices.

Lesson 1 • Topic Modeling and Theme Extraction

Applies AI to identify latent topics and recurring themes across large text corpora. Students interpret and label AI-generated topics for business relevance.

Lesson 2 • Automated Text Summarisation and Reporting

Applies AI to condense large document sets into accurate, audience-appropriate summaries. Students evaluate summary quality and integrate outputs into analytical reports.

Lesson 3 • Sentiment and Opinion Analysis

Uses generative AI to classify sentiment and extract opinion themes from customer text. Students apply results to product, service, and communication improvement decisions.

Lesson 4 • Entity Recognition and Information Extraction

Uses AI to extract named entities, relationships, and key facts from unstructured documents. Students build extraction pipelines for contracts, reports, and communications.

Lesson 1 • Measuring ROI of AI Analytics Initiatives

Establishes metrics and measurement frameworks to quantify AI's analytical value. Students build business cases that connect AI investment to measurable outcomes.

Lesson 2 • Building Automated AI Analysis Pipelines

Constructs automated pipelines that chain AI tasks from data ingestion to insight delivery. Students configure triggers, error handling, and output routing in pipeline design.

Lesson 3 • Mapping and Redesigning Analytics Workflows

Audits existing analytics processes to identify high-value AI integration points. Students produce redesigned workflow maps with AI touchpoints and handoff logic.

Lesson 4 • AI Output Quality Control at Scale

Implements monitoring and validation systems to maintain AI output quality in production. Students design quality gates that catch errors before insights reach decision-makers.