ISLA IPGT 26725

Aprendizagem automática II

• ApresentaçãoPresentation

  The Machine Learning II curricular unit deepens the foundations and methodologies of machine learning, integrating Artificial Intelligence, Data Science and knowledge extraction. It covers the CRISP-DM process, data preparation and transformation, data leakage prevention, model training, evaluation and optimisation, as well as supervised and unsupervised learning techniques. Algorithms such as linear and logistic regression, decision trees, Random Forest, Gradient Boosting, K-Means and hierarchical clustering are explored. The unit adopts a practical approach, focusing on the development and interpretation of integrated Machine Learning solutions in Python, while also addressing model transparency, bias and responsible use.
• ProgramaProgramme

  Course Content Introduction: knowledge extraction; CRISP-DM methodology; formulation of Machine Learning problems; public datasets and application examples. Data preparation and transformation: variable types, exploratory analysis, data cleaning, missing values, outliers, encoding, normalisation, discretisation, feature selection and transformation, pipelines and data leakage prevention. Model training, evaluation and optimisation: training, validation and test sets; holdout and K-fold cross-validation; loss functions; overfitting and underfitting; metrics; imbalanced classes; model selection and hyperparameter optimisation. Supervised learning: linear and logistic regression, decision trees, Random Forest and Gradient Boosting. Unsupervised learning: clustering, K-Means and hierarchical clustering. Interpretation and application: features, predictions, bias, transparency, responsible use and an integrated Python solution.
• ObjectivosObjectives

  To explore the principles and characteristics of machine learning, using methodologies from search, knowledge representation and reasoning, planning, and machine learning. To introduce the concept of artificial intelligence through machine learning. The essential foundations of artificial intelligence in the fields of machine learning and data science. Machine Learning: supervised and unsupervised algorithms.
• BibliografiaBibliography

  Applied Machine Learning Solutions with Python: Production-ready ML Projects Using Cutting-edge Libraries and Powerful Statistical Techniques, Siddhanta Bhatta Machine Learning: Master Supervised and Unsupervised Learning Algorithms with Real Examples (English Edition) by Dr Ruchi Doshi (Author), Dr Kamal Kant Hiran (Author), Ritesh Kumar Jain (Author), Dr Kamlesh Lakhwani (Author).  
• MetodologiaMethodology

  The teaching methodology combines the presentation and discussion of course content with a strong practical component through exercises and the development of Python applications. Innovative and active learning approaches are used, including Project-Based Learning and Problem-Based Learning, encouraging students to solve real-world problems using datasets, experimentation and evidence-based decision-making. The course also incorporates learning by doing, case studies, collaborative learning and the guided use of Artificial Intelligence tools to support programming, analysis and interpretation of results. Moodle is used to provide resources, activities and continuous learning support, fostering autonomy, critical thinking and the practical application of knowledge.
• LínguaLanguage

  Português
• TipoType

  Semestral
• ECTS

  6
• NaturezaNature

  Mandatory
• EstágioInternship

  Não
• AvaliaçãoEvaluation

  Descrição dos instrumentos de avaliação (individuais e de grupo) - testes, trabalhos práticos, relatórios, projetos... respetivas datas de entrega/apresentação... e ponderação na nota final.

   

  Descrição	Data limite	Ponderação
  Teste de avaliação	13-01-2027	60%
  Trabalho de avaliação individual	07-01-2027	30%
  Assiduidade e participação	---	10%

   

  Adicionalmente poderão ser incluídas informações gerais, como por exemplo, referência ao tipo de acompanhamento a prestar ao estudante na realização dos trabalhos; referências bibliográficas e websites úteis; indicações para a redação de trabalho escrito...