Multi-objective Genetic Algorithm for Variable Selection in Multivariate Classification Problems: A Case Study in Verification of Biodiesel Adulteration

Abstract

This paper proposes multi-objective genetic algorithm for the problem of variable selection in multivariate calibration. We consider the problem related to the classification of biodiesel samples to detect adulteration, Linear Discriminant Analysis classifier. The goal of the multi--objective algorithm is to reduce the dimensionality of the original set of variables; thus, the classification model can be less sensitive, providing a better generalization capacity. In particular, in this paper we adopted a version of the Non-dominated Sorting Genetic Algorithm (NSGA-II) and compare it to a mono-objective Genetic Algorithm (GA) in terms of sensitivity in the presence of noise. Results show that the mono-objective selects 20 variables on average and presents an error rate of 14%. One the other hand, the multi-objective selects 7 variables and has an error rate of 11%. Consequently, we show that the multi-objective formulation provides classification models with lower sensitivity to the instrumental noise when compared to the mono-objetive formulation.

Measurement of Diesel Cetane Number Using Near Infrared Spectra and Multivariate Calibration

“…Rapid analysis of diesel fuel properties is of great importance. Recently, near infrared spectroscopy (NIRS) has become an effective method for rapid and real-time analysis of fuel properties [4][5][6]. The objectives of this study were (1) to investigate the feasibility of using NIR spectroscopy to predict the diesel cetane; (2) to obtain the optimal preprocessing methods for the diesel cetane; (3) to achieve the best calibration models after the comparison of PLS and LS-SVM models; and (4) three inputs of Least Squares Support Vector Machine(LS-SVM) calibration method are compared for measurement accuracy of diesel cetane number, namely principal components (PCs), latent variables (LVs), and effective wavelengths (EWs).…”

Method for inverse synthetic aperture radar imaging of space debris using improved genetic algorithm

“…a group of chromosomes, to simulate the evolution in the ecosphere, and the scale of the population P equals to the number of chromosomes in the population. In the proposed method, we set [21].…”

“…In this paper, we apply bit variation in gene mutation, and the mutation probability is set to be p for each bit. The gene mutation operation includes: (a) set the mutation genes at probability [21] for each chromosome; (b) reverse the selected mutation genes, i.e. change 0 to 1 and change 0 to 1, to generate a new offspring, i.e.…”

“…In the propagation process, the gene crossover also occurs, which means that a couple of chromosomes may exchange parts of them, and this procedure includes the following steps: Step 1: Select the chromosome to mate according to the preset mating rate (0.1–0.2) [21]. Step 2: For each couple of chromosomes, select some genes as the crossover points. Step 3: Exchange the selected genes for the couple of chromosomes. Mutation and crossover produce new genes and increase the diversity, which enable the GA to find the global optimal solution with sufficient generations.…”

A Hybrid Algorithm for the Reliability Evaluation Models of Chemical Systems

“…Nowadays, the combination of GA and BP is widely used to gain better prediction results . GA is used to determine initial weights and thresholds of BP neural network, to improve accurate prediction and gain fast convergence …”