Caricamento in corso...
Application of Machine Learning and Data Science
01WPYZV
A.A. 2026/27
Course Language
Inglese
Degree programme(s)
Minor Program D-Tea: Digital Technologies For Environmental Applications - Torino
Course structure
| Teaching | Hours |
|---|---|
| Lezioni | 27 |
| Esercitazioni in laboratorio | 33 |
Lecturers
| Teacher | Status | SSD | h.Les | h.Ex | h.Lab | h.Tut | Years teaching |
|---|---|---|---|---|---|---|---|
| Lombardo Luca | Ricercatore a tempo det. L.240/10 art.24-B | IMIS-01/B | 12 | 0 | 0 | 0 | 1 |
Co-lectures
Context
| SSD | CFU | Activities | Area context | CEAR-01/A CEAR-02/A CEAR-04/A IMIS-01/B |
1 1 1 3 |
|
|
|---|
2026/27
In this framework, the present course—the third in the minor—introduces the pillar of digital analysis by equipping students with the fundamental concepts and tools needed to analyze large datasets using machine learning techniques.
In this third course, students will collaborate more deeply, and will achieve knowledge construction, via group work and integrated mini-projects carried out in discipline-focused laboratories that include the analysis of real-world data.
This course consists of two parts.
The first part aims to provide students with the basic concepts of machine learning and data science that are necessary for processing and analyzing experimental data provided by sensors, especially in the field of environmental monitoring.
Through theoretical and practical lectures, students will learn how to implement simple Python applications for data processing based on machine learning techniques using widespread open-source tools (scikit-learn/TensorFlow). Emphasis will be given to the concept of data quality and its effects on result reliability and uncertainty.
During laboratory sessions, students will have the possibility to develop simple machine learning applications for processing prebuilt datasets and to actively analyze the achieved results.
The second part is about application.
Students will work on real datasets collected form real case study or real experiments relating to environmental data and analyse them with the techniques learnt in the first part.
D-TEA is composed of four courses delivered across two semesters in a sequential (non-parallel) format. This structure is designed to ensure a clear and coherent learning pathway.
The linear organization serves two main purposes:
• Focus and consolidation: Students from diverse disciplinary backgrounds can concentrate on one course at a time, promoting deeper understanding.
• Progressive development of competencies: The minor follows a deliberate progression—from measurement and process understanding, to digital analysis, and finally to interdisciplinary application—ensuring curricular coherence and enabling students to effectively integrate their heterogeneous skills.
In this framework, the present course—the third in the minor—introduces the pillar of digital analysis by equipping students with the fundamental concepts and tools needed to analyze large datasets using machine learning techniques.
In this third course, students will collaborate more deeply, and will achieve knowledge construction, via group work and integrated mini-projects carried out in discipline-focused laboratories that include the analysis of real-world data.
This course consists of two parts.
The first part aims to provide students with the basic concepts of machine learning and data science that are necessary for processing and analyzing experimental data provided by sensors, especially in the field of environmental monitoring.
Through theoretical and practical lectures, students will learn how to implement simple Python applications for data processing based on machine learning techniques using widespread open-source tools (scikit-learn/TensorFlow). Emphasis will be given to the concept of data quality and its effects on result reliability and uncertainty.
During laboratory sessions, students will have the possibility to develop simple machine learning applications for processing prebuilt datasets and to actively analyze the achieved results.
The second part is about application.
Students will work on real datasets collected form real case study or real experiments relating to environmental data and analyse them with the techniques learnt in the first part.
At the end of this course, students are expected to be able to:
• Explain the basic concepts of machine learning and data science and the algorithms of the most common neural networks and classifiers.
• Explain the importance of the quality of the input data and understanding its effects on the achieved results.
• Analyze the achieved results and performance with full awareness.
• Implement simple Python applications based on machine learning techniques for processing sensor data, especially in the environmental monitoring field.
• Assess the role of spatial dependence in machine learning models and the main methodological strategies to identify and mitigate biases and errors induced by the geospatial structure of the data.
• Interpret the results of machine learning applied to environmental data
• Contribute to solving an environmental problem by interpreting large datasets in areas such as remote sensing (e.g., pattern detection), sanitary engineering (e.g.,), and hydraulic engineering (e.g., early failure detection and leakage management)
At the end of this course, students are expected to be able to:
• Explain the basic concepts of machine learning and data science and the algorithms of the most common neural networks and classifiers.
• Explain the importance of the quality of the input data and understanding its effects on the achieved results.
• Analyze the achieved results and performance with full awareness.
• Implement simple Python applications based on machine learning techniques for processing sensor data, especially in the environmental monitoring field.
• Assess the role of spatial dependence in machine learning models and the main methodological strategies to identify and mitigate biases and errors induced by the geospatial structure of the data.
• Interpret the results of machine learning applied to environmental data
• Contribute to solving an environmental problem by interpreting large datasets in areas such as remote sensing (e.g., pattern detection), sanitary engineering (e.g.,), and hydraulic engineering (e.g., early failure detection and leakage management)
Basics of mathematical analysis, physics, computer science, probability theory, statistics, and metrology. Knowledge of the Python programming language.
Basics of mathematical analysis, physics, computer science, probability theory, statistics, and metrology. Knowledge of the Python programming language.
First part (30 hours)
Introductory notes on machine learning: application fields, learning, and inference.
Data pre-processing; normalization; data augmentation; training, validation and testing; under-fitting, over-fitting and bias; dimensionality reduction; performance metrics.
Considerations on the effects of data quality and uncertainty on results.
Classification techniques: regressors, trees, and clustering.
Neural networks: structure and typology, shallow and deep learning, back propagation, Stocastic Gradient Descent, ADAM.
Notes on convolutional neural networks, image processing, and generative AI.
Machine learning and the Python language: scikit-learn and TensorFlow.
Development of simple applications based on machine learning techniques.
Implementation of simple applications based on machine learning techniques for processing data from sensors.
Second part (30 hours)
The geomatics application will address the analysis of geospatial and remotely sensed data through machine learning techniques, with specific emphasis on spatially induced bias and errors arising from the spatial structure of the data. A real-data case study will be presented.
The hydraulic engineering part will address the analysis of time-series of hydraulically relevant quantities (e.g., flow rates) to early detection of failures and leakage management.
The sanitary engineering part will address anomaly detection and fault diagnosis in wastewater treatment operations, with emphasis on sensor drift, fouling, calibration loss, and process upsets (e.g., toxic shocks or filamentous bulking indicators). Students will implement unsupervised and semi-supervised approaches (clustering, isolation methods, autoencoder-based indicators) on multivariate time-series, and will discuss how false alarms, delayed labels, and non-stationarity affect reliability in plant decision support.
First part (30 hours)
Introductory notes on machine learning: application fields, learning, and inference.
Data pre-processing; normalization; data augmentation; training, validation and testing; under-fitting, over-fitting and bias; dimensionality reduction; performance metrics.
Considerations on the effects of data quality and uncertainty on results.
Classification techniques: regressors, trees, and clustering.
Neural networks: structure and typology, shallow and deep learning, back propagation, Stocastic Gradient Descent, ADAM.
Notes on convolutional neural networks, image processing, and generative AI.
Machine learning and the Python language: scikit-learn and TensorFlow.
Development of simple applications based on machine learning techniques.
Implementation of simple applications based on machine learning techniques for processing data from sensors.
Second part (30 hours)
The geomatics application will address the analysis of geospatial and remotely sensed data through machine learning techniques, with specific emphasis on spatially induced bias and errors arising from the spatial structure of the data. A real-data case study will be presented.
The hydraulic engineering part will address the analysis of time-series of hydraulically relevant quantities (e.g., flow rates) to early detection of failures and leakage management.
The sanitary engineering part will address anomaly detection and fault diagnosis in wastewater treatment operations, with emphasis on sensor drift, fouling, calibration loss, and process upsets (e.g., toxic shocks or filamentous bulking indicators). Students will implement unsupervised and semi-supervised approaches (clustering, isolation methods, autoencoder-based indicators) on multivariate time-series, and will discuss how false alarms, delayed labels, and non-stationarity affect reliability in plant decision support.
Additional material for this course is available through a COIL (Collaborative Online International Learning) framework developed within the Unite! Seed-Funded Project DTSWM. In particular, five DTSWM modules are relevant to this course and can be used to deepen knowledge or address potential gaps , or differentiate the learning path:
• Basics of Data Science
• Data-Driven Modeling
• ML & DL fundamentals
• Data analysis and Technological applications
• Advanced course on programming Python
Inclusion of this COIL and formal recognition is to be discussed with the lecturers.
Additional material for this course is available through a COIL (Collaborative Online International Learning) framework developed within the Unite! Seed-Funded Project DTSWM. In particular, five DTSWM modules are relevant to this course and can be used to deepen knowledge or address potential gaps , or differentiate the learning path:
• Basics of Data Science
• Data-Driven Modeling
• ML & DL fundamentals
• Data analysis and Technological applications
• Advanced course on programming Python
Inclusion of this COIL and formal recognition is to be discussed with the lecturers.
The course is delivered in a blended format, with 50% asynchronous online activities to help students balance the additional workload of the minor with other study commitments. Theory components are offered mostly online, while laboratory sessions are conducted primarily in person, typically in interdisciplinary teams. Labs focus on modelling selected physical processes to deepen students’ understanding of the key dynamics introduced during the theoretical lessons.
First part
Theoretical lectures (15 hours)
Practical lectures (3 hours)
Laboratory sessions (12 hours): Four laboratory sessions during which students will be guided towards the application of the concepts explained during the theoretical lessons.
Second part
For each topic of application
Theoretical lectures (3 hours) during which the application and the dataset is explained
Laboratory sessions (7 hours) during which teamswork on the analysis of datasets
The course is delivered in a blended format, with 50% asynchronous online activities to help students balance the additional workload of the minor with other study commitments. Theory components are offered mostly online, while laboratory sessions are conducted primarily in person, typically in interdisciplinary teams. Labs focus on modelling selected physical processes to deepen students’ understanding of the key dynamics introduced during the theoretical lessons.
First part
Theoretical lectures (15 hours)
Practical lectures (3 hours)
Laboratory sessions (12 hours): Four laboratory sessions during which students will be guided towards the application of the concepts explained during the theoretical lessons.
Second part
For each topic of application
Theoretical lectures (3 hours) during which the application and the dataset is explained
Laboratory sessions (7 hours) during which teamswork on the analysis of datasets
Slides and notes provided by lecturers.
Slides and notes provided by lecturers.
Slides; Dispense;
Lecture slides; Lecture notes;
Modalita di esame: Accertamento (esame senza voto);
Exam: Check;
...
Achievement of the more theoretical learning outcomes will be verified through quizzes and tests administered during the course.
Practical skills developed in the lab will be assessed through in-lab discussions and presentations and a concise written report summarizing the work.
Gli studenti e le studentesse con disabilita o con Disturbi Specifici di Apprendimento (DSA), oltre alla segnalazione tramite procedura informatizzata, sono invitati a comunicare anche direttamente al/la docente titolare dell'insegnamento, con un preavviso non inferiore ad una settimana dall'avvio della sessione d'esame, gli strumenti compensativi concordati con l'Unita Special Needs, al fine di permettere al/la docente la declinazione piu idonea in riferimento alla specifica tipologia di esame.
Exam: Check;
Achievement of the more theoretical learning outcomes will be verified through quizzes and tests administered during the course.
Practical skills developed in the lab will be assessed through in-lab discussions and presentations and a concise written report summarizing the work.
In addition to the message sent by the online system, students with disabilities or Specific Learning Disorders (SLD) are invited to directly inform the professor in charge of the course about the special arrangements for the exam that have been agreed with the Special Needs Unit. The professor has to be informed at least one week before the beginning of the examination session in order to provide students with the most suitable arrangements for each specific type of exam.