Benchmarking of Data Preprocessing Methods for Machine Learning-Applications in Production

Procedia CIRP, vol. 104, pp. 50–55

Abstract

The application of machine learning (ML) is becoming increasingly common in production. However, many ML-projects in production fail due to poor data quality. To increase the quality, data needs to be preprocessed. Hundreds of methods exist for data preprocessing (DPP) that are selected manually depending on use-case requirements. For these reasons, DPP is currently performed unstructured and accounts for 80 % of ML-projects’ duration. Thus, we introduce a structured DPP-approach, in which DPP-methods are recommended based on production use-case requirements by benchmarking identified DPP-methods according to ML-model performance on five data sets. The approach is validated through two new use-cases.

Authors 3

1. Maik Frye corresponding Aachen Fraunhofer Institute for Production Technology IPT

   Fraunhofer Institute for Production Technology IPT

   Affiliation as printed

   Fraunhofer Institute for Production Technology IPT, Steinbachstraße 17, 52074, Aachen, Germany

2. Johannes Mohren Aachen Fraunhofer Institute for Production Technology IPT

   Fraunhofer Institute for Production Technology IPT

   Affiliation as printed

   Fraunhofer Institute for Production Technology IPT, Steinbachstraße 17, 52074, Aachen, Germany

3. Robert Heinrich Schmitt Aachen Fraunhofer Institute for Production Technology IPT Laboratory for Machine Tools WZL

   RWTH Aachen University · Fraunhofer Institute for Production Technology IPT

   Affiliation as printed

   Fraunhofer Institute for Production Technology IPT, Steinbachstraße 17, 52074, Aachen, Germany

   Laboratory for Machine Tools WZL RWTH Aachen University, Cluster Production Engineering 3A 540, Aachen 52074, Germany