Machine Learning Foundations: A Case Study Approach Course Syllabus

Overview (80-120 words) describing structure and time commitment.

Module 1: Welcome

Estimated time: 3 hours

• Introduction to machine learning and its business impact
• Overview of tools like Python, Jupyter Notebook, and Turi Create
• Preview of case study-driven learning structure

Module 2: Regression: Predicting House Prices

Estimated time: 3 hours

• Introduction to regression and its use in predicting house prices
• Feature selection, model training, and evaluation
• Implementation using real datasets

Module 3: Classification: Analyzing Sentiment

Estimated time: 3 hours

• Basics of classification with a focus on sentiment analysis
• Text feature extraction and Naive Bayes classification
• Evaluation of prediction accuracy

Module 4: Retrieval: Finding Similar Documents

Estimated time: 3 hours

• Introduction to similarity-based search
• Document representation and nearest neighbor methods
• Use cases in recommendation and content discovery

Module 5: Recommender Systems: Recommending Products

Estimated time: 3 hours

• Collaborative filtering and matrix factorization
• Building recommendation models
• Evaluation metrics for recommender systems

Module 6: Deep Learning: Searching for Images

Estimated time: 3 hours

• Intro to deep learning and neural networks
• Image classification and feature extraction
• Image similarity and search systems

Module 7: Summary and Review

Estimated time: 2 hours

• Recap of key concepts and models
• Guidance on advancing further in ML
• Final quiz and peer review

Prerequisites

• Familiarity with basic programming concepts in Python
• Basic understanding of data structures and file handling
• Access to a computer for installing Python and Turi Create

What You'll Be Able to Do After

• Understand real-world applications of machine learning
• Distinguish between regression, classification, clustering, and recommendation systems
• Apply machine learning techniques using Python and Turi Create
• Evaluate model performance using appropriate metrics
• Build end-to-end ML applications from data preprocessing to deployment