Data Science and Ethics

Course objectives

Given for granted some basic and indispensable goals (knowledge and understanding in the field of studies; ability to apply knowledge and understanding; capability of critical analysis; ability to communicate about what has been learned; skills to undertake further studies with some autonomy), the course intends to attain the following specific objectives: study and acquisition of data analysis techniques based on Artificial Intelligence systems; ethical aspects and privacy preservation in data analysis.

Channel 1
RICCARDO ROSATI Lecturers' profile

Program - Frequency - Exams

Course program
Introduction to data analysis: Exploratory data analysis (EDA); linear regression; clustering (k-means algorithm); data standardization and dimensionality reduction (PCA) Privacy protection: anonymization, pseudo-anonymization, differential privacy, federated learning Bias and fairness in data and machine learning: techniques for measuring bias Explainability and interpretability of data and AI models: interpretable models, symbolic vs. numerical, explanations in symbolic approaches (ontologies), explanations in deep learning models (LIME, SHAP); mechanistic interpretability Data authenticity: distinguishing human data from artificial data; synthetic data and training of Large Language Models
Prerequisites
No prerequisites.
Books
Lecture notes distributed by the teacher.
Frequency
No attendance obligation.
Exam mode
The exam consists of a written test and a written paper (thesis). Optionally, a practical project may be completed, which exempts students from taking the written exam.
Lesson mode
Traditional face-to-face lectures.
MASSIMO MECELLA Lecturers' profile
  • Lesson code10603311
  • Academic year2025/2026
  • CoursePhilosophy and Artificial Intelligence
  • CurriculumSingle curriculum
  • Year3rd year
  • Semester2nd semester
  • SSDING-INF/05
  • CFU6