IMPORT

Research project as part of Software Campus

Improvement of Data Analysis with Data Reduction Techniques

Description

Almost every business sector today needs knowledge, which is gained from data, in order to identify problems, trends and opportunities, as well as to react accordingly to these new insights. However, automatic methods are not yet suitable for recognizing the relevance of the data for a beneficial analysis, since there is a significant domain dependency. Concepts such as visual analytics and self-service business intelligence already aim to give a domain expert more control over this process. But they either solve a specific problem or follow predefined analysis paths on the basis of high-quality data.

For a generic approach Data Mashup tools are often used. These allow for free combination of data sources and operations through an intuitive graphical interface. In particular, these tools are suitable for specifying analysis processes with focus on rapid exploration of data.
But with an increasing number of data sources, it is increasingly complex to select the appropriate data sources.

The goal of this project is to develop methods to support a domain expert in explorative analysis. This includes a pre-selection of data sources, the relief of routine tasks and interaction concepts in the area of data preparation.

This image shows Michael Behringer

Michael Behringer

Dr. rer. nat.

Researcher

To the top of the page