Towards a Process Model to Enable Domain Experts to Become Citizen Data Scientists for Industrial Applications

It is often a problem to combine domain knowledge and data science knowledge in applications of industrial data analytics. Data scientists usually spend a lot of time to understand the domain to develop an application while domain experts lack the skills to interpret results of underlying mathematic...

Full description

Saved in:
Bibliographic Details
Published in2022 IEEE 5th International Conference on Industrial Cyber-Physical Systems (ICPS) pp. 1 - 6
Main Authors Merkelbach, Silke, Von Enzberg, Sebastian, Kuhn, Arno, Dumitrescu, Roman
Format Conference Proceeding
LanguageEnglish
Published IEEE 24.05.2022
Subjects
Adaptation models
citizen data scientist
Conferences
Data models
Data science
domain experts
Home appliances
Mathematical models
process model
Training
Online AccessGet full text
DOI10.1109/ICPS51978.2022.9816871

Cover

More Information
Summary:It is often a problem to combine domain knowledge and data science knowledge in applications of industrial data analytics. Data scientists usually spend a lot of time to understand the domain to develop an application while domain experts lack the skills to interpret results of underlying mathematical models. This leads to difficulties when adapting to changes, handling issues and transfer to similar scenarios, and thus to a lack of acceptance of data analytics applications in industrial companies. Based on the Cross Industry Standard Process for Data Mining (CRISP-DM), we propose a novel process model which integrates training of domain experts to enable them to become citizen data scientists to independently develop and implement data analytics applications. We qualitatively evaluated our process model on a storage location assignment problem in the warehouse of a manufacturer of high-end domestic appliances.
DOI:10.1109/ICPS51978.2022.9816871