Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Model-based data exploration

: Kobialka, Hans-Ulrich; Paurat, Daniel; Schrader, Lisa

presentation urn:nbn:de:0011-n-5256223 (428 KByte PDF)
MD5 Fingerprint: 3f2675204b982a2012b81e5f8e6c6681
Created on: 18.1.2019

Fulltext (PDF; )

Valenzuela, O.:
ITISE 2018, International Conference on Time Series and Forecasting. Proceedings of Papers. Vol.2 : 19-21 September 2018, Granada, Spain
Granada: Godel Impresiones Digitales, 2018
ISBN: 978-84-17293-57-4
International Conference on Time Series and Forecasting (ITISE) <2018, Granada>
Conference Paper, Electronic Publication
Fraunhofer IAIS ()
root cause analysis; failure prediction; offset printing machine; data quality; data labeling

Data exploration is an approach of visually exploring data in order to understand the characteristics of the dataset. As both size and complexity of datasets increase substantially, data scientists take less look at the data directly but conduct experiments by training models and assess the outcome when applying these models on test data. We denote the use of ML models to experimentally obtain insights into the data at hand as model-based data exploration and show some examples from a recent industrial project.