Ateneo - UPO

The course has the objective of introducing the statistical methods for the extraction of information from huge and complex datasets, as those commonly provided by modern instrumentation in laboratories. Several multivariate methods will be described from the theoretical-practical point of view: data pretreatment, pattern recognition methods, classification and regression methods.

Notes and other material provided by the teacher.

The course aims to provide students with solid knowledge of the most common tools of multivariate statistics to prepare the future doctor of Chemical Sciences to face the world of work in the analysis of complex or characteristic datasets of the normal problems encountered in the laboratories. The student will acquire critical sense skills, refine the ability to choose the best experimental strategy and manage complex problems in the technical-scientific field. The course also aims to develop the ability to learn new multivariate analysis techniques independently.
Communication skills: acquiring and knowing how to use an appropriate chemical lexicon in relation to the topics addressed in the course.
Learning skills: the student will be guided to the solution of a proposed case study.

None

Lectures, Powerpoint presentations, manual and computer exercitations, case studies.

The learning during the course will be evaluated by manual and computer exercitations and case studies that will be proposed to the students.
Students with physical disabilities, Learning Disabilities or Special Education Needs can request specific services and tools via the Staff Sviluppo e Coordinamento Carriere e Servizi alle Studentesse e agli Studenti, consulting the University webpage: https://www.uniupo.it/en/services/servicesstudents-physical-or-learning-disabilities.
Students with disabilities, learning disabilities or special education needs, once they have contacted the University Staff, can refer to the tutor in charge of the course to define the examination modalities, concerning academic aspects.

Written exam consisting in two parts: a) 8 multiple choice questions and 8 open questions about the theoretical aspects of the presented methods; b) report on the results of statistical processing carried out on a set of data provided by the teacher, with written commentary on the results obtained.

Statistical methods for the extraction of information from huge and complex datasets (spectroscopic, environmental, etc.). The methods include: data pretreatment (scaling, non linear transformations, missing values, spectral data treatment); clustering techniques (gerarchical, K-means, fuzzy methods), pattern recognition methods (Principal Component Analysis, Multidimensional Scaling), regression methods (calibration theory, Multiple Linear Regression, Partial Least Square, Principal Component Regression, Ridge, variable selection), classification methods (NMC, LDA, QDA, RDA, KNN, Ranking-PCA, PLS-DA, SIMCA, variable selection), artificial neural networks (back-propagatioon, Kohonen, counter - propagation) and genetic algorithms. All lessons have computer sessions with the analysis of real data with dedicated chemometric software.

Knowledge and understanding
- solid knowledge of the theoretical and theoretical / practical bases of the most modern techniques of multivariate data analysis (pattern recognition, classification, regression, non-linear methods)
- knowledge of the main software for multivariate data analysis and how to solve a case study and present it

Ability to apply knowledge and understanding
- know how to apply, through dedicated software, the data analysis techniques seen in class for the solution of proposed case studies;
- know how to compare different methods;
- know how to draw up a technical-scientific report on the analysis of data

Communication skills
- know how to draw up a technical-scientific report on the analysis of data
- acquire and know how to use an appropriate vocabulary in relation to the topics addressed in the course.

Autonomy of judgment
- ability to choose the most suitable approach to the solution of a proposed case study
- being able to critically compare different methods.

Learning ability
- ability to use the study material independently to solve proposed case studies