Visual Analytics for Big Data
Big Data is a key component of modern science and technology. Virtually any modern application field is exposed to the challenge of gathering, analyzing, and making sense of increasingly large, quickly-changing, unstructured, and hybrid amounts of data.
Visual analytics is a key component of this sense-making process. It encompasses a set of techniques aimed at representing big data in suitable ways, depicting this data visually, and allowing users to interactively explore the depiction (visualization) so as to find answers to their questions or find novel insights in the data.
The aim of this course is to provide an in-depth coverage of Visual Analytics techniques aimed at the exploration of Big Data. The course covers the following learning aims:
- explaining what Visual Analytics is and how it relates to other types of data visualization;
- representing Big Data efficiently and effectively for visual analysis purposes (tables, time series, multidimensional data, data attribute types, unstructured data, trees, graphs);
- coverage of several information visualization techniques (color mapping, charting, table lenses, scatterplots, parallel coordinates, multidimensional projections, node-link layouts, treemaps, graph bundling);
- coverage of several practical visual analytics tools;
- visual encoding rules (visual variables, perception, inverse mapping);
- visual design guidelines (linked views, interaction, color usage, annotations, managing visual clutter, ink-data ratio, visual minimalism);
- applying Visual Analytics in real-world problem solving (illustrated by several concrete problem scenarios from the industry).
The targeted audience includes students and (young) researchers working in a data-intensive context, who are interested to learn and use information visualization and Visual Analytics to present and explore large data collections. The course does not cover classical scientific visualization; for that topic, a separate course is available.
The students taking this course should
- have a background in calculus, linear algebra, and statistics (differentiation, integration, matrix/vector operations, first-order differential equations, histograms/distributions);
- have a good background in Computer Science (general data structures and algorithms). Programming skills in a mainstream programming language (e.g. Java, C, C++, Python) are important. Knowledge of computer graphics is advised. Knowledge of scientific visualization is helpful, but not mandatory.
More information on the course is available via the links in the left sidebar.