Chemometrics

Code

Course teacher

Associate teachers

Status of the course

Course objectives

Introduce students to the importance of the use of mathematical and statistical methods for processing experimental data, conduct multi-variety data, and planning experiments. Allow them to work on computers and familiar with standard software packages (MS Excel, Wolfram Mathematica, MatLab, Statistica).

Course enrolment requirements and entry competences required for the course

Completed appropriate undergraduate

Learning outcomes expected at the level of the course (4 to 10 learning outcomes)

1. Define the data distribution.
2. Apply second statistical hypothesis tests in chemistry.
3. Use third method of exploration data on real chemical systems.
4. Know the fourth design an experimental procedure.
5. Put the methods of modeling and optimization, and extract the useful information to know.
6. Know calibrate the analytical system, process measurement signal in order to obtain useful information

Course content broken down in detail by weekly class schedule (syllabus)

Week 1: Introduction to chemometrics. Types of experimental data. The relation between the experimental data, information and knowledge. Seminar: Basic statistical concepts in MS Excel
Week 2: Basic Statistics in chemometrics. Probability. The distribution of the data. Descriptive statistics. The accuracy and precision. Seminar: Basic statistical concepts in Wolfram Mathematica
Week 3: Tests hypotheses. Parametric tests. Significance tests - t-test, F-test, ANOVA test for normality of distribution. Seminar: Basic statistical concepts in Statistics
Week 4: The one-factor analysis of variance. Multi-factor analysis of variance. Seminar: Significance tests in MS Excel
Week 5: Experimental design and optimization. Seminar: Tests of significance in Wolfram Mathematica.
Week 6: The quality of analytical measurement - assessment of variability, comparative tests, measurement uncertainty. Seminar: Tests of significance in Statistics
Week 7: Regression analysis, least squares method: linear models, tests of significance of regression parameters. Seminar: Regression analysis in MS Excel
8th week: Exploratory Data Analysis. A complex pattern. Identifying the sample. Methods for identification of the sample with and without an external teacher. Rotation. Seminar: Regression analysis in Wolfram Mathematica.
Week 9: Principal Component Analysis. Covariance matrix. Eigenvalues and eigenvectors. The principles of reducing the number of dimensions. Seminar: Regression analysis in statistics.
10th week: Hierarchical cluster analysis. Distance and similarity. Single, full and centroid connections. Dendrograms. Seminar: Analysis of the main components in Wolfram Mathematica
11th week: Classification. Linear and nonlinear model of classification. Method K-nearest neighbors. Method independent modeling class analogy. Seminar: Principal Component Analysis in Statistics
12th week: Signal processing. Signal detection, limit of detection, limit choices and limit of quantification. Scaling. Filling. Averaging. Filtering. Leveling. Multi-point sampling. Fourier transformation. Modulation signal. Derivatives signal. Decompression. Seminar: Fourier Transform in Wolfram Mathematica
13th week: Optimization. The functions of the evaluation criteria. Making decisions based on multiple criteria. Pareto optimality. Derringer function. Seminar: Fourier Transform in MatLab
14th week: Algorithms for optimization. Simplex. Genetic algorithms. Basic principles, purpose and usage examples. Seminar: Linear and nonlinear models of classification in Statistics
15th week: molecular modeling. Optimization of the structure. Calculation of descriptors. Linking the physical and chemical properties of the structural properties of the molecules. Seminar: Algorithms for optimization in Statistics

Format of instruction:

Student responsibilities