Processability Analysis using Principal Component Analysis and Support Vector Machine
Yixin Zhang · ERA: Education and Research Archive (University of Alberta) · 2014
The obtained model developed outperforms the existing linear and logistic prediction methods in terms of content prediction error. As the proof of concept, the methodology is applied to an oil sands processing dataset created using an artificial model with such variables as bitumen content and fines content of ores, along with the processing variables such as pH and temperature.