K-Rank: An Evolution of Y-Rank for Multiple Solutions Problem

Abstract

Y-rank can present faults when dealing with non-linear problems. A methodology is proposed to improve the selection of data in situations where y-rank is fragile. The proposed alternative, called k-rank, consists of splitting the data set into clusters using the k-means algorithm, and then apply y-rank to the generated clusters. Models were calibrated and tested with subsets split by y-rank and k-rank. For the Heating Tank case study, in 59% of the simulations, models calibrated with k-rank subsets achieved better results. For the Propylene / Propane Separation Unit case, when dealing with a small number of sample points, the y-rank models had errors almost three times higher than the k-rank models for the test subset, meaning that the fitted model could not deal properly with new unseen data. The proposed methodology was successful in splitting the data, especially in cases with a limited amount of samples.

Rapid Quantification of Lipids in Microalgae Scenedesmus sp. Using Fluorescence

“…If the test metrics are consistent with the training metrics, the ability of the model to deal with new data is confirmed, and the models can be considered more robust and reliable. The train/test split was based on the y-rank methodology: the samples were sorted in ascending order of the output of interest (in this case, the percentage of lipid content), and then the samples were allocated to the train and test subsets following the pattern 2 (train) to 1 (test) [24,25]. The split resulted in a training subset with 66% of the total samples (8 samples) and a test subset with 34% (4 samples).…”

Prediction of sulfur content in diesel fuel using fluorescence spectroscopy and a hybrid ant colony - Tabu Search algorithm with polynomial bases expansion

“…The methodology for splitting the data into calibration (cal), validation (val), and testing (test) subsets is the one implemented by Santos et al [45]. This methodology is especially useful when dealing with multiple solutions problems: situations where standardizes combinations of the input variables can yield the same output y.…”