Objectives and competences
The aim of the course is to train students in development of menthods for BigData analytics, implementation of advanced data fusion and structuring techniques, analysis of approaches to knowledge discovery and categorization and comparison of BigData visualisation techeques.
Content (Syllabus outline)
• Introduction: definition of BigData, key data sources, analysis of current trends, and contemporary challenges.
• Architectural components of systems for BigData analytics (MapReduce framework, Hadoop and Spark) and data representation in NoSQL databases (column, document, key-value and graphs).
• Data structuring: DAM data model and cache-oblivious implementation of data structures (key-value tables, graphs and spatially embedded graphs).
• Analysis of graphs and complex networks: graph search and graph query algorithms and metrics of complex network metrics (node degree, centrality, communities).
• Descriptive analysis and association rules with adaptation of apriori algorithm for large data volumes.
• Inductive data analysis (central tendency, dispersion, and Bayesian theorem).
• Time series analysis: definition of regression models, autoregression, and ARIMA model.
• Survival analysis: proportional hazard regression, parametric survival analysis, and Kaplan Meier analysis.
• Recommendation systems: alternating least squares in singular value decomposition.
• Social sensing and social network analysis: relation neighbourhood classifier, relation logistic regression, and collective interfacing.
• BigData visualization: force directed graphs, regression, residual, interactive, violin, strip, point, and count plots, heatmaps, pair grids, and facet grids.
Learning and teaching methods
• lectures,
• seminar work
• lab work,
• individual work.
Intended learning outcomes - knowledge and understanding
On completion of this course the student will be able to
• demonstrate knowledge in development of BigData technologies
• remove inconsistencies within the data
• analyse and categorize advanced machine learning approaches
• implement predictive, descriptive, time series, and survival analysis
• design and implement recommendation systems,
• develop and analyse algorithms for assessment of complex networks,
• implement advanced visualisation of BigData.
Intended learning outcomes - transferable/key skills and other attributes
• Communication skills: oral defence of practical exercises and manner of expression at written examination.
• Use of information technology: the use of programming languages, data processing techniques and machine learning.
• Problem solving: implementation of advanced data analysis
Readings
• B. Baesens: Analytics in a big data world: The essential guide to data science and its applications, John Wiley & Sons, New Jersey, USA, 2014.
• A. Bahga and V. Madisetti: Big Data Science & Analytics: A Hands-On Approach, Bahga & Madisetti,Georgia, India, 2016.
• N. Marz and J. Warren: Big Data: Principles and best practices of scalable realtime data systems, Manning Publications Co., New York, USA, 2015.
Additional information on implementation and assessment The exam may be replaced by written midterm examinations in the weight of 50%.
