Chemometrics 0600-S2-CKR-Ch

Lecture content:
Chemometrics is a science that uses numerical, statistical, and symbolic methods to analyze multidimensional chemical data sets, empirical modeling methods, similarity analysis, and classification.
The lecture is intended to present theoretical knowledge on planning experiments, modeling dependencies, advanced statistical methods, and interpretation of the obtained experimental results.

Laboratory program content:
The exercises aim to acquire practical skills enabling the independent use of computer programs in preliminary data processing, modeling relationships, planning experiments, and advanced statistical methods.
In the case of exercises, it is necessary to have the ability to use an MS Excel spreadsheet and knowledge of matrix calculus.

Total student workload

Contact hours with teacher: - participation in lectures – 15 h - participation in exercises – 30 h - consultations with an academic teacher - 10 h Self-study hours: - preparation for exercises – 50 h - preparation for test/examination – 20 hours Altogether: 125 hours: 25 = 5 ECTS

Learning outcomes - knowledge

W1: Knows the methodology, goals, and areas of application of chemometric methods K_W01, K_W04 W2: Knows methods of analysis and control of raw data and their transformation (normalization, standardization) K_W01 W3: Knows various methods of planning and optimizing experiments K_W01 W4: Knows the basic techniques of filtering data K_W01 W5: Knows how to create a single- and multi-dimensional chemometric regression model K_W01 W6: Can describe the algorithms of operation of the most important chemometric methods: principal component analysis, principal component regression, partial least squares, and discriminant analysis K_W01 W7: Knows examples of the practical application of chemometrics in forensics K_W04

Learning outcomes - skills

U1: Can analyze and transform raw data K_U06 U2: Applies in practice selected methods of planning measurements and optimizing experiments K_U01, K_U05 U3: Uses various statistical calculation programs K_U01 U4: Can search for open-source software used in chemometric calculations K_U01 U5: Can construct single- and multi-parameter chemometric regression models and check the significance of variables, model adequacy and prognostic abilities K_U06 U6: Can apply selected data filtering methods K_U06 U7: Can analyze multi-dimensional experimental data using selected chemometric methods and correctly interpret the results K_U01, K_U06, K_U07

Learning outcomes - social competencies

K1: Is aware of the level of his knowledge of chemometrics and the need for continuous training and personal development K_K01 K2: Is aware of the role of chemometrics in forensics K_K01

Teaching methods

Lecture: Expository teaching methods: Informative (conventional) lecture Laboratory: Exploratory teaching methods: Practical, classic problem-solving

Expository teaching methods

- informative (conventional) lecture
- participatory lecture

Exploratory teaching methods

- practical
- laboratory

Prerequisites

MS Excel basic knowledge, mathematics at the secondary school level, and knowledge of matrix calculus

Course coordinators

Piotr Szczepański

Assessment criteria

Assessment methods:
lecture - K_W01, K_U01, K_U06, K_U07
exercises - K_W01, K_U01, K_U06, K_U07
Assessment criteria:
Lecture: written examination, test: fail- 0÷49 %, satisfactory- 50÷60%, satisfactory plus- 61÷65%, good - 66÷75%, good plus- 76÷80%, very good- 81÷100%.
Laboratory: two tests fail- 0÷49 %, satisfactory- 50÷60%, satisfactory plus- 61÷65%, good - 66÷75%, good plus- 76÷80%, very good- 81÷100%.

Lecture: written exam – K_W01, K_U01, K_U06, K_U07

Laboratory: - K_W01, K_U01, K_U06, K_U07

Exam topics:
Selected methods of planning experiments, chemometric modeling (model identification, model significance, model adequacy, significance of variables), data analysis and control (distribution, covariation, data transformation, scaling, linearization), similarity analysis (multidimensional similarity, feature similarity analysis, similarity analysis of objects, reduction of the dimension of the feature space), cluster analysis (graphic methods, dendrogram, k-nearest neighbours method) data filtering methods, principal component analysis (calculation of the main components, determination of the number of main components and their interpretation), principal component regression method (PCR), partial least squares method (PLS), object and feature classification methods.

Practical placement

Not applicable

Bibliography

1. Richard G. Brereton, Chemometrics: data analysis for the laboratory and chemical plant, Wiley, 2003.
2. Practical guide to chemometrics, ed. Paul Gemperline, Taylor&Francis, 2006.
3. Guideline for the use of Chemometrics in Forensic Chemistry, European Network of Forensic Science Institutes Drugs Working Group, 2020.
4. Georgina Sauzier, Simon W. Lewis, Chemometric Methods in Forensic Science, Royal Society of Chemistry, 2023

Additional information

Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system:

Description of 0600-S2-CKR-Ch in USOSweb