Machine Learning for managers

Lecture designed to give business students a practical, non-technical grounding in ML concepts, covering the main types of supervised learning through intuitive real-world examples. The lecture closes with a business-oriented lens on ROI, ethical considerations, and model selection trade-offs, reinforcing that models are decision-support tools rather than replacements for human judgment.

Contents covered:

• Introduction to AI and Machine Learning (definitions and the AI landscape)
• Data formats (tabular, time series, images, videos, text)
• Types of Machine Learning (supervised, unsupervised, reinforcement — with the latter two only briefly mentioned)
• Supervised Learning: Regression (linear regression, cost function, evaluation metrics: MSE, RMSE, MAE, R²)
• Supervised Learning: Classification (threshold-based classifiers, K-Nearest Neighbors, Decision Trees, confusion matrix, precision & recall)
• The Supervised Learning Framework (train/validation/test split, model tuning, deployment)
• Model Comparison & Selection (overfitting vs. underfitting, simplicity vs. performance, interpretability vs. accuracy)
• Practical Business Considerations (data quality, bias, fairness, change management)
• ROI of ML projects (benefits vs. costs breakdown)