Data Preprocessing

Garant předmětu:

Přednášející:

Cvičící:

Předmět zajišťuje:

katedra aplikované matematiky

Anotace:

Students learn to prepare raw data for further processing and analysis. They learn what algorithms can be used to extract parameters from various data sources, such as images, texts, time series, etc., and learn the skills to apply these theoretical concepts to solve a specific problem in individual projects - e.g., parameter extraction from image data or from Internet.

Požadavky:

Fundamentals of statistics, FCD course in data mining.

Osnova přednášek:

1. Data exploration, exploratory analysis techniques, visualization of raw data.

2. Descriptive statistics.

3. Methods to determine the relevance of features.

4. Problems with data ? dimensionality, noise, outliers, inconsistency, missing values, non-numeric data.

5. Data cleaning, transformation, imputing, discretization, binning.

6. Reduction of data dimension.

7. Reduction of data volume, class balancing.

8. Feature extraction from text.

9. Feature extraction from documents, web. Preprocessing of structured data.

10. Feature extraction from time series.

11. Feature extraction from images.

12. Data preparation case studies.

13. Automation of data preprocessing.

Osnova cvičení:

1. Assignment of course projects.

2. Consultations.

3. Presentation of course projects.

Cíle studia:

Data preprocessing is crucial for successful data processing and takes a lot of time - usually more than the data processing itself. Knowledge of algorithms for extraction of parameters from various data sources is a fundamental part of knowledge engineering,

Studijní materiály:

1. Pyle, D. ''Data Preparation for Data Mining''. Morgan Kaufmann, 1999. ISBN 1558605290.

2. Guyon, I., Gunn, S., Nikravesh, M., Zadeh, L. A. ''Feature Extraction: Foundations and Applications (Studies in Fuzziness and Soft Computing)''. Springer, 2006. ISBN 3540354875.

Poznámka:

Information about the course and courseware are available at https://courses.fit.cvut.cz/MIE-PDD/

Další informace:

https://courses.fit.cvut.cz/MIE-PDD/

Pro tento předmět se rozvrh nepřipravuje

Předmět je součástí následujících studijních plánů: