AI for Data Analyst 课程安排

课程大纲

AI Basics
AI Engineering Basics

Goal: Understand what LLMs are, how they work, and how we can talk to them using prompts.

1. What is a Large Language Model (LLM)?

Simple explanation of LLMs

Examples of what they can do: text generation, Q&A, summarization

1. What are Embeddings?

What are embeddings in simple terms (turning words into numbers)

Why embeddings are useful: similarity search, clustering, etc.

1. What is Prompt Engineering?

Why prompts matter

Zero-shot, few-shot, and chain-of-thought examples

Practice: Write good and bad prompts

1. Intro to RAG (Retrieval-Augmented Generation)

What is RAG and why use it

How RAG combines your data with LLM

Hands-on: Build a basic RAG app in Python using LangChain + Streamlit

AI for Data Analysis&Azure AI Services Overview
How AI Can Help with Data

Goal: Learn how to use AI tools to analyse data and create structured outputs.

1. How AI Can Help with Data

Examples: summarize survey results, classify customer feedback, extract insights from text

1. Hands-on Practice

Use AI to extract key insights, categories, or summaries from sample data, and generate structured outputs

Azure AI Services Overview

Goal: Get familiar with Azure AI’s tools and when to use each one.

1. Overview of Azure AI Services

Language: translation, text analytics, summarization

Vision: image analysis, OCR

Speech: speech-to-text, text-to-speech

Azure AI Search: vector search + keyword search

1. Azure AI Foundry (previously AI Studio)

What it is and how it helps build, test, and deploy AI apps

Quick tour of the UI

AI数据分析实操
AI数据分析&Copilot 自动化
Goal:
Master AI-powered data analytics and automate common workflows without writing code, using Copilot and Azure AI Services.
1. Introduction to AI-driven Data Analysis
1. What Can AI Do for Data Analysts?
2. Quick overview: Leveraging AI to automate repetitive analytics tasks
3. Scenarios: Data cleaning, quick data summarization, insights generation
4. Hands-on with Excel Copilot and Power BI Copilot
5. Using natural language to create pivot tables and charts instantly
6. AI-generated data insights and automated summarization
7. Demo: Generating weekly sales reports with a single prompt
2. Copilot for Workflow Automation (Zero-code)
1. Automating Weekly CRM Data Reports
2. Step-by-step demo: Auto-send CRM data to the Sales team's mailbox every week
3. Leveraging Power Automate & Copilot without coding
4. Automatic Email Reply & Customer Support
5. Build an AI-powered email responder that automatically replies to frequent customer queries
6. Demo: Using Copilot and Azure AI to manage inbox and respond to emails efficiently
3. Practical Exercises & Real-world Scenario
Tools Covered:
1. Excel Copilot
2. Power BI Copilot
3. Azure AI Services
4. Microsoft Fabric
5. Power Automate
Data Security&Dynamic Data Masking
Data Security&Dynamic Data Masking

Goal: Understand how to protect data using access controls, masking, and encryption.

1. Role-Based Access Control (RBAC)

What RBAC is and why it's important

Who gets access to what (based on role)

Simple demo: set RBAC in Azure

1. Data Protection Techniques

Data Masking: hide sensitive info

Encryption: protect data at rest and in transit

Hashing: irreversible transformation for password

Talk about best practices in a real-world project