Machine Learning Concepts: Supervised and Unsupervised

In this course, students will learn the fundamentals of Machine Learning, understanding how computers can learn from data without explicit programming. The two main approaches will be presented: Supervised Learning, which uses labeled data to build predictive models with algorithms such as Regression, Decision Trees, and Neural Networks, and Unsupervised Learning, which explores hidden patterns in unlabeled data through techniques like clustering, association rules, and dimensionality reduction.

The module emphasizes practical applications that impact the real world, including Natural Language Processing (NLP) for virtual assistants and chatbots, AI in video games to create intelligent opponents and personalized worlds, and the convergence of robotics and data mining for autonomous manufacturing and Big Data analysis. Students will understand the data quality and interpretation challenges in each approach, highlighting the importance of validation and contextual knowledge.

Finally, future trends and challenges will be discussed, such as hybrid techniques combining supervised and unsupervised methods, reinforcement learning, Explainable AI for transparent decision-making, and Generative AI, which requires human oversight and rigorous regulation to mitigate algorithmic biases. The module reinforces that ethical responsibility is just as essential as technical competence to ensure AI delivers real and social benefits.