The main goal of the course is to introduce students to modern techniques and methods for efficient data analysis, extracting useful information and presenting it in a way that will help the executives of a company to make useful business decisions. In the course will be studied the main techniques of design and development of data warehouses as well as analysis of multidimensional data models. Students will also be introduced to a wide range of data analysis techniques that can be used in understanding business data, extracting knowledge from it and in the decision-making process. Through this course, students are expected to learn techniques that are part of business intelligence and gain important technical skills in business data analysis.
Multidimensional data model, data warehouse architecture, data warehouse design, data export-transformation-loading.
Multidimensional data analysis
OLAP functions, query systems in data warehouses, report generation.
Introduction to recommendation systems
Introduction to the problem and applications of recommendation systems. Presentation of basic techniques for personalized recommendations through content-based approaches, techniques of closest neighbors. User-user collaborative filtering technique, (item-item collaborative filtering) algorithm.
Advanced recommendation techniques
Matrix derivatization methods and hybrid methods of recommendations.
Methods for finding frequently occurring sets, market basket analysis, Apriori algorithm, metrics for evaluating correlation rules.
Exploratory analytics and visualizations
Univariate and bivariate analysis, visualization, histograms, cumulative distribution function, summary statistics, position and dispersion measures, detection of correlations between two variables, alternative mapping methods using diagrams, use of visualization techniques for multivariate data analysis.
Detailed with visualization
Data visualization tools and techniques, data analytics with visualization, Tableau visualization tool, applications in business intelligence, new user interfaces, advanced visualization techniques, original research systems.
Time series analysis
Examples and motivations, trend detection, moving averages, smoothing methods, autocorrelation function.
The role of simulation to extract information from data, Monte-Carlo simulation, use of simulation for cases where analytical modeling is complex, development of models with simulation, validation of models with simulation.
Tags for geographic data, social networks that take into account the position, a combination of spatial, temporal and text data, analytical applications aimed at geographical social content, analytical position on Twitter, Flickr, Foursquare.
- Han J. & Kamber M. (2006): Data Mining: Concepts and Techniques, 2nd Edition, Morgan Kaufmann.
- Jure Leskovec, Anand Rajaraman, Jeff Ullman. Mining of Massive Datasets. Cam-bridge University Press. 2014 (2nd Edition).
- Raymond T.Ng et al. (2013): Perspectives on Business Intelligence. Morgan & Claypool Publishers. Synthesis Lectures on Data Management.
- Philipp K. Janert (2010) Data Analysis with Open Source Tools: A hands-on guide for programmers and data scientists, O’Reilly Media.