Review of Data Mining methods (in SAS Viya) 2400-ZEWW905
During seminars the following essential problems for Data Mining analysis will be considered:
1. Initial data analysis: qualitative data analysis methods, quantitative data analysis methods.
Fuzzy sets and rough sets.
2. Discriminant analysis. Logistic regression. Factor analysis and principal component analysis.
3. Classification and regression trees. Generating fuzzy rules.
4. Cluster analysis. Fuzzy method of cluster analysis. Muti-layer undirectional neural networks. Kohonen neural networks.
5. Genetic algorithms. Chaos theory. Correspondence analysis.
6. Methodology of the SEMMA data mining process. Data mining tools that support the analysis at respective SEMMA stages. Stages in the SEMMA methodology and the structure of data analysis diagrams. General rules for creating a diagram.
7. Preparing data for data mining. Initial analysis of input data. Conducting variable transformations. Solutions for the problem of missing values. Variable selection. Using the Tree Node to determine variables. Selecting variables for the Regression Node.
8. Forecasting methods. Logistic regression, decision trees and neural networks - modelling standards and the analysis of outcomes (default settings of the application). The assessment of forecast accuracy. Generating project reports. Setting parameters and adjusting models for a forecast: logistic regression, decision trees, gradient boosting, random forests and neural networks.
9. Association analysis.
Type of course
Course coordinators
Learning outcomes
After completion of the course students should possess competence in Data Mining analysis using Data Mining models and methods and have the knowledge how to use SAS Viya software
Tuition outcomes:
1) Knowledge
Student has understanding of available Data Mining methods and techniques and knows how to use them for purpose of economic researches.
2) Skills
Student can choose and put to use appropriate methods and techniques for specific applications, if necessary, is able to modify (adapt) the methods and techniques to make them the most efficient and effective when carrying out given case studies.
3) Social competences
Student is aware of the possibilities and advantages of data mining methods and techniques in economic research and the benefits that can be obtained through their use, but also aware of the defects and the limited usefulness of these methods and techniques.
KW01, KW02, KW03, KU01, KU02, KU03, KK01, KK02, KK03
Assessment criteria
Presence during classes is required. The condition to obtain a credit is to prepare and present a review of scientific article using one of methods discussed during the classes (50% of the grade) and complete homeworks (50% of the grade).
Additional information
Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system: