Welcome to Data Processing Compendium – Workflows for Knowledge Exploitation in the Process Industries!¶
The Data Processing Compendium is a catalogue of methods for data processing, so far with a focus on data pre-processing. The compendium is a key result of the AiF research project Knowledge based pre-processing of process data (IGF-Vorhaben Nr. 16901 N). The project has been conducted by AixCAPE® e.V.
Support and feedback by 12 industrial partners from Belgium, France, and Germany is greatly acknowledged.
Overview¶
Algorithms¶
Algorithms such as Backward Difference and Holt-Winters Double Exponential Smoothing are the foundation to solve many data processing tasks. Algorithms can be accessed in three different ways:
- Algorithms: An alphabetical list of algorithms.
- Algorithms by Classification: Algorithms classified by basic properties, for example One-Dimensional Algorithm and Multi-Dimensional algorithm.
- Algorithms by Tool: Algorithms grouped by implementing tools, for example Excel and Matlab.
Single Steps¶
Single steps provide a task-oriented view on data processing. They are arranged in a hierarchy of tasks, for example Single Steps → Data Cleaning → Handling Improper Values → Reconstructing Improper Values. The lowest level in the hierarchy contains the actual single steps, typically identified by an algorithm, for example Reconstructing Improper Values with Local Centered Moving Median.
- Single Steps: Hierarchy of single steps.
- Single Steps by Classification: Single steps classified by basic properties, for example Outlier-Resistant Step.
Reference Processes¶
Reference processes are best practice procedures for more complex tasks from literature or provided by our project partners.
- Reference Processes: List of reference processes.
Feedback¶
Users of the compendium are invited to send feedback and suggestions for improvement to info(at)aixcape.org (replace (at) with @).
Project Partners¶
- aiXtrusion GmbH
- Bayer Technology Services GmbH (now Bayer AG)
- BLANCO CS GmbH + Co KG
- Computing in Technology (CiT) GmbH
- engineo GmbH
- Evonik Industries AG
- HiTec Zang GmbH
- IPCOS NV
- LeiKon GmbH
- m2p-labs GmbH
- Total S.A.
- umesoft GmbH
Acknowledgment¶
Das IGF-Vorhaben Nr. 16901 N der Forschungsvereinigung DECHEMA e.V., Theodor-Heuss-Allee 25, 60486 Frankfurt am Main, wurde über die AiF im Rahmen des Programms zur Förderung der industriellen Gemeinschaftsforschung und -entwicklung (IGF) vom Bundesministerium für Wirtschaft und Energie aufgrund eines Beschlusses des Deutschen Bundestages gefördert.
The IGF project 16901 N of the research association DECHEMA e.V., Theodor-Heuss Allee 25, 60486 Frankfurt am Main, Germany, has been funded via AiF within the program for industrial joint research and development (Industrielle Gemeinschaftsforschung und -entwicklung, IGF) by the German Federal Ministry of Economics and Energy based on a resolution of the German Federal Diet.
License¶
The Data Processing Compendium is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
