99107 - SCIENZA DEI DATI PER L'AMBIENTE

Learning outcomes

Knowledge of the main topics of analysis and modelling of univariate, bivariate and multivariate data.

Course contents

Prerequisites

Fundamentals of statistics and probability theory.

Program

Elements of data engineering

• Big data
• Collection of raw data
• Data transformation
• Data sharing (database, data lake, data warehouse)

Descriptive statistics

• Representation of data in summary form (tables, graphs)
• Sorting and distribution of data
• Covariance, covariance matrices, and correlation
• Dimensionality reduction of data (singular value decomposition, principal component analysis, factor analysis)
• Recognition of implicit relational structures among data

Learning methods

• Parametric supervised learning: linear regression
• Non-parametric supervised learning: classification
• Non-parametric unsupervised learning: clustering
• Machine Learning (ML)
• Artificial Neural Networks (ANN)
• Genetic Algorithms (GA)

Elements of inferential statistics

Fundamentals of scripting

Anonymous statistical survey

Once two-thirds of the lessons have been completed, a statistical survey will be conducted to gather students’ opinions about the course in order to make it more effective. Reference sites:

• https://opinionistudenti.unibo.it
  
• https://val.unibo.it/
• https://val.unibo.it/demo.php
• https://gestioneval.unibo.it

Additionally, please include your personal reflections on the following points:

• Which topics did you find less engaging or relevant?
• Which aspects do you feel deserve deeper exploration?
• Were there any points that seemed unclear or required further explanation?
• What was the most valuable insight or takeaway for you?
• Do you believe any additional content should be included in the course?

Readings/Bibliography

The lecture notes distributed by the instructor through the official teaching materials platform Insegnamenti OnLine is required reading for exam preparation.

To further explore the course content, the following useful links are suggested:

• Introduction to Environmental Data Science
  Jerry Davis, SFSU Institute for Geographic Information Science
• Data Science e Machine Learning: dai Dati alla Conoscenza
  Michele di Nuzzo
• Modern Statistics with R From Wrangling and Exploring Data to Inference and Predictive Modelling
  Måns Thulin
• Data Science. Guida ai Principi e alle Tecniche Base della Scienza dei Dati
  Sinan Ozdemir
• Statistica per Data Science con R
  Enrico Pegoraro
• R for Data Science
  Garrett Grolemund - Hadley Wickham
• Metodi Statistici per la Sperimentazione Biologica
  Alessandro Camussi, Frank Möller, Ercole Ottaviano, Mirella Sari Gorla
  Zanichelli, II edizione, 1995

Teaching methods

The course consists of 6 ECTS divided into two modules:

• Theory module, 4 ECTS
• Laboratory module, 2 ECTS

The lessons lasts 4 hours and include both a theoretical part (lecture and exercises) and a practical computer session using the students' computers, so as to become familiar with some of the methods underlying the subject.

The software used for this course is R [https://cran.r-project.org/], and all students are required to install it on their computers prior to the start of the lessons.

The program is free and available for the main operating systems:

• Linux
• MacOS (also install XQuartz [https://www.xquartz.org/] to enable the graphical interface)
• Windows

It is also required to install the latest version of:

• Pandoc [https://pandoc.org/installing.html]

Optional installation:

• RStudio [https://posit.co/download/rstudio-desktop/]

Given the types of activities and teaching methods used, participation in this course requires all students to complete Modules 1 and 2 in e-learning mode via the following link:

https://www.unibo.it/en/services-and-opportunities/health-and-assistance/health-and-safety/online-course-on-health-and-safety-in-study-and-internship-area

Assessment methods

The assessment is aimed at verifying the acquisition of both the theoretical knowledge and the practical skills expected. The final grade reflects an evaluation of the content demonstrated during the exam.

The student must present a program for analyzing a dataset, agreed upon with the instructor, which will serve as the basis for discussing the topics covered in class.

The use of artificial intelligence tools is permitted for generating content that will be subject to critical discussion and evaluation during the examination.

The exam lasts 30–45 minutes. 

As a guideline, the following evaluation criteria are provided:

• Failing

  • Incomplete knowledge of the subject
  • Lack of orientation within the topics
  • Inappropriate language

• Passing

  • Minimal knowledge of the subject
  • Analytical ability emerges only with the instructor’s help
  • Barely appropriate language

• Adequate

  • Good memorized knowledge of the subject
  • Fair argumentative ability
  • Correct language

• Excellent

  • Clear understanding and mastery of the subject
  • Excellent ability to elaborate and argue
  • Specific and appropriate language

https://corsi.unibo.it/magistrale/AnalisiGestioneAmbiente/qualita-corso/@@esami-voto-medio

Students who require compensatory tools due to temporary or permanent disabilities, or specific learning disorders (SLD) may contact the appropriate University office well in advance:

• https://site.unibo.it/studenti-con-disabilita-e-dsa/en

The office will be responsible for proposing any necessary adjustments, which must be submitted at least 15 days before the exam date for the lecturer's approval. The lecturer will assess their appropriateness in relation to the learning objectives of the course.

Teaching tools

Blackboard (lectures and exercises), video projector, internet connection.

Computational laboratory practicals

Sustainable Development Goals

Quality Education (4) The course develops advanced competencies in quantitative data analysis, the informed use of digital tools, and the critical interpretation of scientific information, contributing to the education of qualified professionals in the environmental sector.

Climate Action (13) The statistical and machine learning methods covered in the course can be applied to the analysis of climate, environmental, and monitoring data, supporting the assessment, prediction, and management of phenomena related to climate change.

Life on Land (15) Quantitative analysis of environmental, ecological, and territorial data provides a valuable tool for understanding and supporting the sustainable management of terrestrial ecosystems and biodiversity.

Office hours

See the website of Assimo Maris