2026-04-14T02:50:37Zhttps://recercat.cat/oai/requestoai:recercat.cat:2117/4466502026-01-13T04:21:17Zcom_2072_1033col_2072_452950
RECERCAT
author
Karna, Ashutosh
author
Gibert, Karina
2024
This study focuses on developing an intelligent decision support system (IDSS) that helps a human operator make data-driven decisions. To put IDSS in production, it is necessary to develop two additional components: one oriented to recognize the cluster of new data and the other a knowledge-based resulting from the interpretation of clusters and further association of actions to each cluster, constituting a knowledge base with the alerts and recommendations associated to every profile. Bootstrap-CURE technique is used to handle the initial component, whereas a meta-clustering framework is suggested for interpreting the clusters and providing recommendations. A detailed strategy is presented for handling a print job, examining patterns, and executing actions through IDSS, thus improving predictive accuracy and operational efficiency. Two distinct machine learning models were developed, one to detect the operational mode and another to choose the best meta-cluster for the type of printing jobs and detained steps are provided for implementing the recommendations.Peer ReviewedPostprint (published version)
http://creativecommons.org/licenses/by-nc/4.0/ Open Access Attribution-NonCommercial 4.0 International
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Clustering
3D printers
Pattern discovery
Bootstrap clustering
Intelligent decision support system
From clustering to intelligent decision support system: An application to 3D printing
Conference report