Classification of iron ores by laser-induced breakdown spectroscopy (LIBS) combined with random forest (RF)

Abstract

Laser-induced breakdown spectroscopy (LIBS) integrated with random forest (RF) was developed and applied to the identification and discrimination of ten iron ore grades. The classification and recognition of the iron ore grade were completed using their chemical properties and compositions. In addition, two parameters of the RF were optimized using out-of-bag (OOB) estimation. Finally, support vector machines (SVMs) and RF machine learning methods were evaluated comparatively on their ability to predict unknown iron ore samples using models constructed from a predetermined training set. Although results show that the prediction accuracies of SVM and RF models were acceptable, RF exhibited better predictions of classification. The study presented here demonstrates that LIBS–RF is a useful technique for the identification and discrimination of iron ore samples, and is promising for automatic real-time, fast, reliable, and robust measurements.