Human Aware Data Analysis – Visual & Guided Data Analytics

26.08.2020 | Sprache: english

Research and technology trends in Big Data and AI

We are in the process of preparing our fourth COMET application and are looking for partners to jointly explore the potential of these key technologies:
Learn more about the advantages and opportunities of a COMET partnership and secure your access to funded top-level research.

Target group:

Researchers, engineers, and everyone else facing the task of exploring, navigating, and analyzing data using interactive tools and systems. While some of the presented techniques are generally applicable, others may target a particular application domain, such as industry, media, or health.


The recent advances in technology resulted in an enormous increase in personal and industrial data. Although this data obtains valuable information that can be used to better support humans and machines, it is a tedious and time-consuming task to extract and use this information to gain valuable insights and draw correct conclusions. Although there exist several methods that can help to effectively mine the data (e.g., AI methods) they often require users to possess certain skills. For example, AI methods require users to select suitable features for data profiling, define the correct parameter space for the learning process, etc. Unfortunately, average users do not bring adequate experience and have serious difficulties when using such complex methods. More importantly, users rarely know which method to use to e.g., analyze outliers, trends, distribution, and the root cause of certain events. To support users when analyzing their data, we propose methods which (i) recommend a sequence of interactions (previously provided by the experts) that shows which (AI) methods are best suited for the current analytical process and (ii) displays the final outcome of the analysis visually using an appropriate chart selected by the system with regard to visual encoding rules and perceptual guidelines. Yet, our assistance does not end here. Out methods further provide means to help users locate the visual patterns (i.e., outliers, trends, correlations, etc.) and navigate them to these interesting areas on the visualizations.

After the event you will know:

About advanced methods for interactive data analysis using visual tools and systems, e.g.,

  • Personalized visualization and user interfaces

  • Guided data analytics

  • AI to support users in analytical processes

  • Domain-specific visual analytics techniques (e.g. for health or media)

  • Big data visualization (with focus on time-series data)

  • Gathering user feedback to improve AI algorithms

You will also learn about

  • Previous success stories on interactive data exploration and visual analytics

  • Ongoing projects and future research plans


Vedran Sabol

Vedran Sabol

Research Area Manager Knowledge Visualization

Belgin Mutlu

Belgin Mutlu

Tobias Schreck

Tobias Schreck

Professor, TU Graz

Christian Partl

Christian Partl

Senior Researcher Knowledge Visualization