48
Data Processing Systems
REGGIO DI CALABRIA
Overview
Date/time interval
Syllabus
Course Objectives
The course aims to provide students with the theoretical and practical knowledge necessary to understand and apply the main techniques of Artificial Intelligence.
At the end of the course, the student is able to:
- understand the fundamental principles of Artificial Intelligence, acquiring the skills needed to analyze data in different application contexts;
- use techniques and tools to develop, train, and evaluate AI models, with attention to good implementation practices and correct interpretation of results.
Course Prerequisites
It is important that students have knowledge of the following topics:
- basic programming concepts (Foundations of Computer Science);
- basic notions of algebra.
During the lectures, the fundamental concepts will be recalled when necessary to ensure full understanding of the topics covered.
Teaching Methods
The course consists of 40 hours of lectures and 8 hours of practical computer exercises.
The lectures are aimed at introducing the theoretical foundations of machine learning.
The practical exercises are carried out on computers and are dedicated to the application of theoretical concepts through laboratory activities that involve practical exercises in Python.
Assessment Methods
The assessment consists of a written exam (duration: 2 hours) and the development of a project.
During the written exam, students will be required to answer theoretical questions related to the main machine learning concepts covered during the course.
The project consists of developing a machine learning application based on a dataset agreed upon with the instructor. Students must document the choices made during the different stages of the work (data preprocessing, model selection and training, evaluation metrics used) and present the obtained results in a short report.
The final evaluation will take into account both the theoretical understanding of the concepts and the ability to apply machine learning techniques in practice, as well as methodological correctness and clarity of presentation.
Evaluation Criteria
30 – 30 with honors (Excellent)
The student demonstrates a complete and detailed knowledge of the topics covered, using appropriate terminology and explaining the main concepts of the discipline accurately. The student shows an in-depth understanding of machine learning methods, clearly distinguishing between fundamental elements and supporting aspects. They are able to integrate theoretical and practical knowledge, compare different methodological approaches, justify project choices, and critically analyze the obtained results. The presentation is clear, rigorous, and well structured.
28 – 29 (Very Good)
The student demonstrates a thorough knowledge of the subject and a solid understanding of the main machine learning concepts. They are able to correctly apply the studied techniques and interpret the results obtained in the project. The presentation is clear and well organized, although minor inaccuracies or limited depth of analysis may be present.
25 – 27 (Good)
The student has a good knowledge of the main topics of the course and is able to apply basic machine learning techniques to the proposed project. Key concepts are understood, although they may not always be explained with full precision or completeness. The presentation is generally clear, but sometimes schematic or not fully developed.
22 – 24 (Fair)
The student demonstrates an overall adequate knowledge of the topics covered, although with some gaps or uncertainties. They are able to apply the fundamental machine learning concepts, although with limited autonomy or some inaccuracies in the interpretation of the results. The presentation is correct but not always fully structured.
18 – 21 (Satisfactory)
The student demonstrates an essential knowledge of the fundamental machine learning concepts, although with gaps and some difficulties in explaining and applying the studied methods. The project and written exam show a basic understanding of the subject and an elementary application of the techniques covered.
Fail
The student does not demonstrate an adequate knowledge of the fundamental topics of the course. The written exam and/or the project show significant errors, major gaps, or an inability to correctly apply machine learning concepts, and therefore the minimum requirements to pass the exam are not met.
Texts
Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow
Course handouts
Jupyter notebooks provided during the course
Contents
Course Program
The course is divided into two modules.
The first module (3 CFU) introduces the basic concepts of machine learning and the classical techniques of predictive modeling. The topics are as follows:
(1 CFU)
- Introduction to Artificial Intelligence and Machine Learning
- Overview of Python and key libraries (NumPy, Pandas, Scikit-learn)
- Dataset preparation and preprocessing
(2 CFU)
- Linear and polynomial regression
- Supervised classification
- Logistic regression
- Support Vector Machines
- Decision Trees and Random Forests
- Clustering and basic techniques of unsupervised machine learning
The second module introduces neural networks and their advanced applications for processing images, sequences, and text. The topics covered include:
(1 CFU)
- Fundamentals of artificial neural networks (perceptron, feed-forward networks, backpropagation)
(1 CFU)
- Convolutional Neural Networks (CNN) for image recognition
- Basic notions of adversarial attacks
(1 CFU)
- Recurrent Neural Networks (RNN, LSTM, GRU) for sequential data
- Introduction to Natural Language Processing (NLP)
- Large Language Models (LLMs) and their applications
Expected Results
Knowledge and understanding
After passing the exam, the student knows and understands the theoretical foundations of Artificial Intelligence, the main Machine Learning and Deep Learning algorithms, data preprocessing and analysis techniques, and model evaluation methodologies. The student is also able to understand the technical documentation of the most widely used tools and frameworks in the field.
Ability to apply knowledge and understanding
Upon completion of the course, students will be able to apply their theoretical knowledge to the design and implementation of machine learning and deep learning models for real-world problems. They will also be able to adapt existing models to new application contexts, evaluating their limitations and potential.
Making judgments
After passing the exam, the student will be able to independently evaluate different methodological and algorithmic alternatives when designing AI models. They will be able to compare different approaches based on application requirements, performance, interpretability, computational complexity, and ethical considerations related to data usage.
Communication skills
During the phases of analysis, design, and development of AI models, the student will be able to communicate effectively with non-technical stakeholders, translating problems expressed in natural language into technical specifications and proposing understandable solutions, highlighting advantages, limitations, and implications of the chosen methodologies. The student will also be able to present and justify the results obtained through clear visualizations and explanations.
Learning skills
During the course, students will learn to consult and interpret technical documentation and online resources, mainly in English. This will enable them to continuously update their skills independently and to apply this study methodology to new and emerging technologies in the field of Artificial Intelligence.
More information
Teams code: gynvusz
I invite students to join the Team to access the teaching materials.