A data mining process for building recommendation systems for agricultural machines based on big data
Vorschaubild nicht verfügbar
ISSN der Zeitschrift
42. GIL-Jahrestagung, Künstliche Intelligenz in der Agrar- und Ernährungswirtschaft
Gesellschaft für Informatik e.V.
There is a potential expansion in the agricultural machinery industry by using the collected data from different years. Big data is already being used in other industries like e-commerce to improve decision-making processes. There are several existing process models to lead through the generic processes of data mining. The common factor between the process models that have attained dominant public position is that they are domain-agnostic frameworks. This paper proposes a method to extend the CRoss-Industry Standard Process for Data Mining (CRISP-DM) to focus on the agricultural domain and give guidelines on how to handle and structure the agricultural data and processes to reach defined data mining goals. The paper provides a walk-through for a use case to build a recommendation system.