Knowledge Discovery in Databases

Completion:examination (verbal)
Type of
Hour/semLecturesSem. ExercisesLab. exercisesComp. exercisesOther
Guarantee:Zendulka Jaroslav, doc. Ing., CSc., DIFS
Lecturer:Zendulka Jaroslav, doc. Ing., CSc., DIFS
Faculty:Faculty of Information Technology BUT
Department:Department of Information Systems FIT BUT
Advanced Database Systems (PDB), DIFS
Knowledge Discovery in Databases (ZZN), DIFS
Substitute for:
Knowledge Discovery in Databases (ZZD), FIT
Learning objectives:
  To deepen students' knowledge in the field of knowledge discovery in databases and other data sources (KDD) with special focus on theoretical foundations of the used techniques, algorithms and models.
  1. The deepening of basics in KDD - basics of methods of data preprocessing (statistics quantities used in data summarization, approaches to data cleaning, transformation and reduction), basics of data warehousing, basic methods and algorithms of mining frequent items and patterns and association rules (Apriori algorithm, FP-tree, multi-level association rules, mining multidimensional association rules from relational databases), basic methods and algorithms of classification (decision tree, Bayesian classification, using neural networks, SVM) and prediction (linear and nonlinear regression), basic methods and algorithms of cluster analysis (distance of data, partitioning methods, hierarchical methods, CF-tree, density-based methods, grid- and model-based methods).
  2. Advanced data mining techniques - advanced techniques of data mining in 'classic' data sources, mining in data streams, time series and sequences, mining in biological data; mining in graphs, multirelational data mining, mining in object, spatial and multimedia data, mining in text, mining on the Web.
Knowledge and skills required for the course:
  Students should have basic knowledge in statistics, database systems, information theory, machine learning, neural networks. It is assumed that they have passed some subject on KDD.
Learning outcomes and competences:
  Students get a broad, yet in-depth overview of the field of data mining and knowledge discovery. They get a deeper view mainly in the field related to the topic of their thesis.
Syllabus of lectures:
  1. Data preprocessing. 
  2. Data warehousing.
  3. Asociation analysis.
  4. Classification and prediction.
  5. Cluster analysis.
  6. Advanced data mining in 'classic' data sources.
  7. Mining in data streams.
  8. Data mining in time series and sequences.
  9. Mining in biological data.
  10. Data mining in graph structures.
  11. Mining in object, spatial and multimedia data.
  12. Text mining and Web mining.
  13. Mining moving object data.
Syllabus - others, projects and individual work of students:
  1. Reading up and treatment of a selected topic concerning knowledge discovery in a field related to the student's PhD thesis.
Fundamental literature:
  • Han, J., Kamber, M.: Data Mining: Concepts and Techniques. Third Edition. Elsevier Inc., 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.
Study literature:
  • Bishop, CH. M.: Pattern Recognition and Machine Learning. Springer, 2006, 738 p. ISBN 978-0-387-31073-2.
  • Aggarwal, Ch.C. (ed.): Data Streams: Models and Algorithms. Advances in Database Systems. Springer, 2006, 358 p. ISBN 0387287590.
  • Papers in journals and conference proceedings (including those in ACM Digital library, IEEE Digital library and other electronic sources).
Controlled instruction:
  Consultations, elaboration of a given topic, written report and presentation on the final seminar.
Progress assessment:
  Control questions during consultations.