Currículo

Data Science and AI for Sustainability DSAIS

Contextos

ECTS

3.0 (para cálculo da média)

Objectivos

The curriculum for the Doto Science ond AI for Sustainability course is structured to align with learning outcomes, ensuring students acquire essential data analysis skills tailored to sustainability challenges. The course begins with an introduction to Python, its main libraries, and environments such as Jupyter Notebook and Spyder, while incorporating AI tools like Large Language Models (LLMs} and Generative AI to support programming and problem-solving. The integration of libraries like Pandas, NumPy, Matplotlib, and Seaborn strengthens proficiency in data manipulation and visualization, with a focus on sustainability data. From there, the course explores data importing from various formats and graphical representations to analyze environmental, social, and economic datasets. Subsequent sessions address data cleaning, transformation, and integration techniques, including handling missing data, outlier detection, and normalization. The curriculum also emphasizes exploratory data analysis, covering univariate and multivariate statistics applied to sustainabi/ity metrics such as carbon emissions, resource efficiency, and biodiversity trends. Finally, practical case studies with real-world sustainability data and group projects prepare students to apply their acquired skills to address pressing environmental and societal challenges.

Programa

1. Introduction to Python, Data Science & AI Tools • Python installation • Python libraries • Anaconda environments: Jupyter Notebook, Spyder • Introduction to Data Science • Overview of AI tools: Large Language Models (LLMs) and Generative AI for coding assistance • Using Al-driven tools like ChatGPT, LLaMA and Claude 2. Programming with Python • Pandas, Numpy, Matplotlib, and Seaborn libraries • Tuples and lists • Conditional and loop structures • Functions 3. Data Importing, Representation, and Visualization • Data types • Importing data in CSV, Excel, and text formats • Graphical data representations 4. Data Cleaning, Transformation, and Integration • Handling missing data • Detecting and correcting outliers • Data errors and duplicates • Data encoding • Data standardization and normalization • Data discretization • Data integration 5. Exploratory Data Analysis • Descriptive statistics • Correlation and association measures between numerical variables • Correlation and association measures between categorical variables 6. Case Studies • Case studies with cross-sectional data • Case studies with time-series data • Group project development and discussion with real data.

Método de Avaliação

Teaching Methodologies • Theoretical Lectures and Practical Labs: Engaging presentations complemented by visual resources and real-world examples to clarify complex topics. Students work on programming exercises using Python, focusing on real-world data analysis problems. • Group Projects: Students collaborate in teams to develop a software solution or analysis project, fostering peer learning and teamwork. • Case Studies: Analysis of real-world case studies showcasing successful applications of programming in data analysis, discussing challenges and solutions • Peer Review and Presentations: Students present their group projects, receiving feedback from peers and instructors, promoting a constructive learning environment. The assessment for Data Science and AI or Sustainability consists of: • Regular period: Group Project (60%) and Final Exam (40%). • Repeated period: Appeal Exam (100%). The Final and Appeal Exams are two-hour written exams. The Group Project (4-6 people) must be emailed to the instructors and discussion group, with a printed copy submitted before or during the presentation. The report (in PPT format) should have 1!r-15 slides, including cover, tables, figures, and table of contents. Presentations are 20 minutes, followed by a 10-minute discussion. Grades are based on the written report, oral presentation, and discussion. Students must justify their methods and model choices. Uneven contributions may result in different group member grades. Questions and discussion quality also influence the group grade. 8. Demonstra~ao da coerencia das metodologias de ensino com os objetlvos de aprendizagem da unidade curricular. Demonstration of the coherence between the teaching methodologies and the learning outcomes. Campo alfanumerico {3.000 carateres). Esta Unidade Curricular esta organizada de forma a potenciar o envolvimento ativo dos alunos no processo de aprendizagem. Neste sentido, sera solicitado os alunos a realiza~ao de um trabalho de grupo no qual deverao demonstrar que adquiriram algumas das competencias que supostamente adquiriram ao Iongo do semestre. 0 modelo de avalia~ao foi desenhado para apoiar este modelo de aprendizagem. This Course Unit is organized in order to enhance the active involvement of students in the learning process. In this sense, the students will be required to complete a group project assignments where they are expected to demonstrate they have acquired the skills they are supposed to have acquired during the semester. The assessment model was designed to support this learning model.

Carga Horária

Bibliografia

Principal

• Generative AI with Python and Tensorflow 2, Packt Publishing.: Babcock, J. and Raghav Bali, R. 2021
• Data Science with Python, Packt Publishing Ltd: Chopra, R. and England, A. 2019
• Python e Explora~oo de Dodos, Edi~oes Sflabo, Usboa.: Caiado, J. 2024

Secundária

Disciplinas de Execução

2025/2026 - 2 Semestre