Machine based approach for structure based zeolite classification

• Main Author: Karan, Braba
• Other Authors: Su Haibin
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/67301
• Institution: Nanyang Technological University

This project aims at comparing two distinct classifiers and their ability to accurately classify zeolites to its framework structure. The software chosen for this project is WEKA. This is one of the software used by students from computational science in consolidating variables and coming up with predictions. The objective of this project is to accurately classify a zeolite into its framework type based on its largest cavity diameter. If proven viable, this would allow scientists to save time in analysis such as SEM, TEM and XRD to classify a zeolite into its respective framework type. The results of the classification type are based entirely on the computational algorithm set by the classifier therefore; it reduces the human errors that might be involved with this research. As more and more zeolite structures are being identified and with them, billions of dollars affecting the industry, this research is important as it saves time which translates to money in the industrial world.