Lectures (topics): 1. The process of acquiring knowledge - history, definition of goals, review of methodologies. 2. Dividing of datamining tasks, introducing typical tasks. 3. Data preparation, data comprehension, dataset description, data matrix preparation, data selection and cleaning, data source design and aggregation, type homogeneity, data formatting. 4. Association algorithms - search for association rules, model Apriori, Carma, implied statistics, prediction model. 5. Fraud detection - fraud, credit risk, behavioral scoring for risk assessment of repaid loans. 6. Classification algorithms as predictive tools based on historical data. Decision trees, C & RT algorithms, C5.0, CHAID, QUEST. Conversion of tree to rules, pruning trees. 7. Discrimination analysis - classification of cases into classes, scoring. 8. Segmentation algorithms - Uncovering unusual structures in data by applying grouping algorithms K-Means, Two Step, Anomaly. 9. Fundamentals of neural networks for processing of categorized and numerical variables, use in cases where classical linear methods do not provide the expected results. 10. Analysis and prediction of time series using DM models, preparation of data, addition of missing values, differences, seasonal differences, moving averages and medians, smoothing of time series. 11. Modeling and evaluating solutions, putting DM solutions into practice, including scoring processes in the company's decision-making workflow. 12. Web Mining 13. Text Mining. 14. New Trends in Data Mining. Seminars (themes): 1. Basics of work With IBM SPSS Modeler, data import, data comprehension. 2. Data matrix preparation, data manipulation, data audit. 3. Open-source datamining tools. 4. Association Rules for Basket Analysis. 5. Detection of fraudulent requests - fraud. 6. Prevention of money laundering. 7. Classification tasks. 8. Customer Segmentation. 9. Credit risk scoring, evaluation in datamining tasks. 10. Prediction of customer behavior. 11. Defense of semester papers.
|
-
HAN, Jiawei. a Micheline. KAMBER, 2012. Data mining: concepts and techniques. 3rd ed.. Burlington, MA: Elsevier., 2012. ISBN 9780123814791.
-
HOFMANN, Markus a Ralf. KLINKENBERG. RapidMiner: Data Mining Use Cases and Business Analytics Applications.. Florida: Taylor & Francis Group., 2013. ISBN 9781482205497.
-
PETR, Pavel. Metody Data Miningu.. Pardubice: Univerzita Pardubice, 2014. ISBN 9788073958732.
-
SHMUELI, Galit, Peter C. BRUCE, Mia L. STEPHENS a Nitin R. PATEL. Data mining for business analytics: concepts, techniques, and applications in JMP Pro. 1.. Canada: WILEY, 2016. ISBN 978-1-118-87743-2.
-
WENDLER, Tilo a Sören GRÖTTRUP. Data mining with SPSS modeler: theory, exercises and solutions. 1. Switzerland: Springer, 2016. ISBN 978-3-319-28707-2.
-
WITTEN, I. H. a Frank EIBE. Data mining: practical machine learning tools and techniques:Fourth Edition. Cambrige, 2017. ISBN 9780128042915.
|