Knowledge Discovery in Databases

• 39 hrs lectures
• 13 hrs projects

• 51 pts final exam (written part)
• 15 pts mid-term test (written part)
• 34 pts projects

• Students get a broad, yet in-depth overview of the field of data mining and knowledge discovery.
• They are able both to use and to develop knowledge discovery tools.

• Student learns terminology in Czech and English.
• Student gains experience in solving projects in a small team.
• Student improves his ability to present and defend the results of projects.

To familiarize students with the methods and algorithms of data modelling for knowledge discovery from it.

Due to the increasing amounts of data currently stored in databases and other data sources, it is necessary to discover some new knowledge, which is not possible to obtain with use of querying the data. Therefore, in connection with the knowledge and skills from the subject UPA  related to the data mining process and to the data preparation before its modelling, it is necessary to get acquainted with the data modelling methods and algorithms. They are based on methods and techniques from various areas, such as statistics and machine learning. 

• Knowledge of the basic steps of the data mining process and methods of data preparation for the step of data modelling (discussed in the subject UPA - Data Storage and Preparation).
• Basic knowledge of probability and statistics.
• Knowledge of database technology at a bachelor subject level. 

• Han, J., Kamber, M.: Data Mining: Concepts and Techniques. Third Edition. Morgan Kaufmann Publishers, 2012, 703 p., ISBN 978-0-12-381479-1.

• Han, J., Kamber, M.: Data Mining: Concepts and Techniques. Third Edition. Morgan Kaufmann Publishers, 2012, 703 p., ISBN 978-0-12-381479-1.
• Han, J., Kamber, M.: Data Mining: Concepts and Techniques. Second Edition. Elsevier Inc., 2006, 770 p., ISBN 1-55860-901-3.  

1. Data Warehouse and OLAP Technology for knowledge discovery.
2. Mining frequent patterns and associations - basic concepts, efficient and scalable frequent itemset mining methods.
3. Multi-level association rules, association mining and correlation analysis, constraint-based association rules.
4. Predictive modelling - basic concepts, classification methods - decision tree, Bayesian classification, rule-based classification.
5. Classification by means of neural networks, SVM classifier, Random forests.
6. Other classification and regression methods. Evaluation of quality of classification and regression.
   Cluster analysis - basic concepts, types of data in cluster analysis.
7. Partitioning-based and hierarchical clustering. Other clustering methods. Evaluation of quality of clustering.
8. Outlier analysis. Mining in biological data.
9. Introduction to mining data stream and time-series.
10. Introduction to mining in sequences, graphs, spatio-temporal data, moving object data and multimédia data. 
11. Text mining.
12. Mining the Web. Process mining.
13. Introduction to big data analytics.

• Working-out a data mining project by means of an available data mining tool.

A mid-term test, formulation of a data mining task, presentation of the project.
Exam prerequisites:
Duty credit consists of working-out the project, defending project results and of obtaining at least 24 points for activities during semester.

• Mid-term written exam, there is no resit, excused absences are solved by the guarantor.
• The formulation of the data mining task in the prescribed term, excused absences are solved by the assistent.
• The presentation of the project results in the prescribed term, excused absences are solved by the assistent.
• Final exam, The minimal number of points which can be obtained from the final exam is 20. Otherwise, no points will be assigned to the student. excused absences are solved by the guarantor.

Duty credit consists of working-out the project, defending project results and of obtaining at least 24 points for activities during semester.