gh-bradleyboehmke-brads-marketplace-course-builder/skills/courses/intro-to-data-mining/course-profile.md

Intro to Data Mining - Course Profile

Course: UC BANA 4080: Introduction to Data Mining with Python Instructor: Brad Boehmke

Audience

Student Level: Undergraduate (juniors/seniors)

Background:

• Business students with foundation in calculus, statistics, and possibly regression
• May have Excel experience and basic VBA exposure
• Variable programming backgrounds - many are complete coding beginners
• Understand business operations, customer behavior, and quantitative thinking
• Know how to think critically but lack experience turning theory into practice with code

Prerequisites:

• Quantitative methods and statistical inference courses
• No prior programming experience required or expected
• Course explicitly designed for beginners

Key Challenge: Bridging the gap between classroom theory and real-world data analysis

Learning Philosophy

Core Approach: Hands-on, immersive learning through doing

Key Principles:

1. Practice over theory - Students learn by working with real, messy datasets
2. Build confidence through action - Focus on getting students comfortable with tools before perfection
3. Close the theory-practice gap - Move from "knowing concepts" to "applying skills"
4. AI as assistant, not autopilot - Use GenAI tools (ChatGPT, Claude, Copilot) to help learn, but emphasize understanding
5. Collaborative learning - Build community; encourage students to help each other
6. Growth mindset - Normalize struggle; coding is a new language that takes time

Teaching Style:

• Conversational, relatable tone (see intro chapter example)
• Use storytelling and scenarios (e.g., "Taylor the intern")
• Address student concerns directly (e.g., "Why learn this when AI exists?")
• Set realistic expectations about difficulty
• Encourage persistence and resilience

Unique Aspects:

• Explicitly addresses role of GenAI in learning process
• Balances AI assistance with foundational skill building
• Uses real-world business contexts students can relate to

Technical Stack

Core Environment:

• Python (chosen for beginner-friendliness + professional power)
• Jupyter Lab/Notebooks (primary development environment)
• Google Colab (cloud-based option for students)
• Quarto (for textbook and slides)
• Virtual environment (venv for package management)

Primary Libraries (Weeks 1-6):

• pandas - data manipulation and DataFrames
• numpy - numerical computation
• matplotlib - basic visualization
• seaborn - statistical visualization

Machine Learning Libraries (Weeks 8-13):

• scikit-learn - all ML models and evaluation

Additional Tools:

• CSV/Excel file handling
• Basic SQL concepts (joins in pandas)
• Git/GitHub for assignment submission

File Formats:

• Quarto markdown (.qmd) for book chapters
• Jupyter notebooks (.ipynb) for examples, labs, homework
• Real datasets (CSV, Excel) in /data/ directory

Content Style

Writing Style:

• Conversational and approachable - Not dry or overly academic
• Student-focused - Addresses "you" directly
• Motivational - Builds confidence, normalizes struggle
• Practical - Always tied to real-world application
• Honest - Acknowledges difficulties, doesn't sugar-coat challenges

Explanation Approach:

1. Start with WHY - Motivate the topic before diving in
2. Use analogies and stories - Make abstract concepts concrete
3. Show, don't just tell - Working code examples over theory
4. Progressive complexity - Start simple, build gradually
5. Address common questions - Anticipate student concerns

Examples:

• Use relatable business scenarios (customer data, marketing analytics, retail transactions)
• Work with messy, real-world datasets (not clean, perfect examples)
• Include visual aids heavily (plots, diagrams, screenshots)
• Provide executable code that students can run and modify

Pedagogical Elements:

• Callout boxes for tips, warnings, reflections, and examples
• Student reflection prompts to encourage metacognition
• Exercises that build on chapter concepts
• Code comments that explain what's happening
• Error messages and debugging guidance

Depth:

• Prioritize intuition over mathematical rigor
• Show code implementation before heavy theory
• Balance "just enough math" with practical application
• Focus on interpretation and application over derivations

Key Topics

Module 1: Python Fundamentals (Week 1)

• Course intro + motivation
• Variables, data types, basic operators
• Why Python? Why not just use AI?
• Setting up environment

Module 2: Jupyter & Data Structures (Week 2)

• Jupyter notebooks and reproducible workflows
• Lists, dictionaries, tuples
• Pandas introduction
• Importing CSV data

Module 3: Data Wrangling (Week 3)

• DataFrame manipulation
• Filtering and subsetting
• Aggregating data
• GroupBy operations

Module 4: Advanced Data Manipulation (Week 4)

• Working with dates and times
• String operations
• Relational data and joins (SQL-style in pandas)
• Merging DataFrames

Module 5: Data Visualization (Week 5)

• Matplotlib basics
• Seaborn for statistical plots
• Exploratory data analysis with visuals
• Best practices for effective visualization

Module 6: Writing Efficient Code (Week 6)

• Control flow (if/else, loops)
• Functions and modularity
• List comprehensions
• Code efficiency and readability

Week 7: Midterm Project

• Application of Modules 1-6
• Work with messy, real datasets
• Open-ended analysis problem

Module 7: Machine Learning Intro (Week 8)

• What is ML and when to use it?
• Train/test split
• Features and labels
• Model building process

Module 8: Regression (Week 9)

• Correlation analysis
• Linear regression with scikit-learn
• Model evaluation (R², RMSE)
• Interpretation

Module 9: Classification (Week 10)

• Logistic regression
• Classification metrics (accuracy, precision, recall, F1)
• Confusion matrices
• When to use classification vs regression

Module 10: Tree-Based Models (Week 11)

• Decision trees
• Random forests
• Feature importance
• Model interpretation

Module 11: Model Optimization (Week 12)

• Feature engineering
• Cross-validation
• Hyperparameter tuning (GridSearchCV)
• Model selection

Module 12: Advanced Topics (Week 13)

• Unsupervised learning (clustering, PCA)
• Deep learning overview
• Introduction to LLMs and GenAI concepts

Week 14: Final Project

• Comprehensive data science project
• Full pipeline from data cleaning to modeling

Assessment Approach

Grading Components:

• Labs - Weekly hands-on activities (Thursdays)
• Homework - Applied assignments (with answer keys for instructor)
• Midterm Project - Comprehensive application of Modules 1-6
• Final Project - End-to-end data science project
• Quizzes - Knowledge checks (materials in /planning/quizzes/)

Student Support:

• Canvas discussion boards for peer collaboration
• Office hours
• Answer keys provided for labs and homework (instructor use)
• Multiple formats (notebook, HTML, PDF) for accessibility

GenAI Policy:

• Encouraged to use ChatGPT, Claude, Copilot as learning aids
• Required to understand code, not just copy it
• Emphasis on using AI to learn, not to avoid learning
• Students asked to reflect on AI tool use and limitations

Project Structure:

• Templates provided for major assignments
• Rubrics included in /planning/projects/
• Real-world datasets required
• Open-ended problems that require creative problem-solving

Content Format

Textbook: Quarto book with modules 1-6 + appendices Slides: Weekly presentations using Quarto + Reveal.js Examples: Numbered sequence of Jupyter notebooks (01-17) Labs: Weekly hands-on activities with answer keys Homework: Individual assignments with solutions in multiple formats Datasets: Real-world data in /data/ directory (retail, airlines, housing, etc.)

Course Materials Repository

All materials maintained in Git repository with structure:

• /book/ - Textbook chapters
• /slides/ - Weekly presentations
• /example-notebooks/ - Companion code examples
• /labs/ - Hands-on activities
• /homework/ - Assignments
• /data/ - Datasets
• /planning/ - Instructor resources (Canvas docs, rubrics, quizzes)