Go back to teaching

FIN 4300J: Principles of Machine Learning

Kean University, Jersey City
#Machine Learning #Python #Statistical Learning #Data Science #Business Analytics

Course Overview

Principles of Machine Learning provides students with a structured introduction to statistical learning, computational modeling, and data-driven decision-making. Students learn how to select appropriate algorithms, implement models in Python, evaluate predictive performance, and interpret model outputs.

The course reviews relevant mathematical foundations—including functions, derivatives, gradients, probability, statistics, loss functions, and optimization—and connects them with practical applications using real-world datasets from business, finance, and related fields.

Topics Covered

  • Python, Jupyter Notebook, NumPy, and pandas
  • Machine Learning Workflows
  • Supervised and Unsupervised Learning
  • Training, Validation, and Test Data
  • Overfitting and the Bias-Variance Trade-Off
  • Linear and Multiple Regression
  • Logistic Regression
  • Linear and Quadratic Discriminant Analysis
  • Naive Bayes and K-Nearest Neighbors
  • Cross-Validation and Model Evaluation
  • Ridge Regression and Lasso
  • Decision Trees, Bagging, Random Forests, and Boosting
  • Support Vector Machines
  • Principal Component Analysis
  • K-Means and Hierarchical Clustering
  • Neural Networks and Deep Learning
  • Model Interpretation, Bias, and Responsible AI

Instructional Approach

The course combines lectures, mathematical analysis, guided Python demonstrations, coding exercises, case studies, collaborative problem-solving, and project-based learning. Concepts are developed through visual explanation, mathematical formulation, algorithmic reasoning, and computational implementation.

Students use Jupyter Notebook or Google Colab to prepare data, implement models, compare algorithms, evaluate performance, and communicate analytical findings. The course is designed to remain accessible to students with limited programming experience while developing the mathematical and computational foundations required for applied machine learning.

Culminating Applied Project

Each student completes an individual machine learning project addressing a clearly defined business or organizational problem. The project incorporates data preparation, exploratory analysis, baseline and comparative models, training and validation strategies, performance evaluation, interpretation, and discussion of limitations, bias, ethics, and responsible use.

Students submit an executable Python notebook and communicate evidence-based recommendations derived from their analysis.

“Machine learning is not simply about producing predictions—it is about understanding data, evaluating evidence, and making responsible decisions.”