Customer behavior analysis
An exploratory data project that turns raw records into clear findings and visual evidence.
AI & Data · AI foundations
AI & Machine Learning Foundations
Understand the machine-learning workflow by preparing data, training models, evaluating results, and shipping a focused AI project.
12 weeksPython fundamentals required5 projects
Course overview
Learn the whole workflow, not isolated tricks.
Build a practical foundation in applied machine learning without skipping the reasoning behind the models. You will explore data, prepare features, train baseline models, evaluate tradeoffs, and communicate results responsibly.
What you will be able to do
Clear outcomes you can demonstrate.
Explore, clean, and prepare datasets for a repeatable ML workflow.
Train baseline supervised-learning models and compare performance.
Choose useful evaluation metrics and identify common model risks.
Package findings into a clear, reproducible portfolio project.
Curriculum
A focused path from foundation to final build.
05Modules
Data and ML foundations
4 topicsProblem types, datasets, features, labels, notebooks, experiments, and responsible use basics.
- ML workflow
- NumPy
- Pandas
- Ethics
Data preparation
4 topicsExploration, missing values, outliers, encoding, scaling, feature selection, and train-test splits.
- EDA
- Cleaning
- Features
- Visualization
Supervised learning
4 topicsRegression, classification, baseline models, trees, ensembles, and practical model selection.
- Regression
- Classification
- Trees
- Ensembles
Evaluation and improvement
4 topicsMetrics, cross-validation, tuning, leakage, imbalance, interpretability, and error analysis.
- Metrics
- Validation
- Tuning
- Interpretability
Applied ML capstone
4 topicsFrame a problem, build a reproducible pipeline, assess limitations, and communicate the result.
- Pipeline
- Experiment
- Reporting
- Demo
Portfolio projects
Build work worth explaining in an interview.
Prediction model
A tested supervised-learning workflow with justified metrics and documented limitations.
Applied ML capstone
A reproducible end-to-end project with a concise report and presentation-ready result.