Advanced Visual Data Analysis

Aim

After completing the course, the student should be able to:

• Examine new, complex data sets and identify relevant features which might be extracted.
• Select and apply algorithmic methods for analysis of large complex data sets to determine valuable results.
• Select and apply well-established visualization methods to display extracted relevant information from such data sets.
• Describe and reflect on visualization techniques (regarding interaction and visual representation) and example tools for special data sets and applications domains based on theory and practice discussed in the course.
• Plan and implement a data analysis approach combining interactive visualization techniques with algorithmic methods in a given context.
• Make well-grounded design choices in the context of various tasks and data constraints and be able to motivate and justify these choices.

Content

This course builds upon the course Information Visualization, with a focus on the data modelling, mining and analysis techniques with are the foundation of modern visual data analysis methodology. Such methods are becoming very important as the scale of data available for analysis expands, leading to the so-called 'big data' problem affecting business, healthcare, government, science and industry.

Topics covered:

Introduction to the field of visual data analysis and relevant concepts, such as visual analytics (VA), knowledge generation model for VA, human-in-the-loop and human-centered analysis.Data aspects that affect analysis.Visualization techniques for spatial, temporal, tree / network, text data and combinations of these.Interaction concept that promotes people-centered analysis.Overview of techniques for automatic analysis support, such as techniques for abstraction, comparison, pattern detection, dimension reduction, text mining.