Evolutionary Approaches to Variable Selection

Evolutionary Approaches to Variable Selection

Marcos Gestal (University of A Coruña, Spain) and José Manuel Andrade (University of A Coruña, Spain)
Copyright: © 2009 |Pages: 8
DOI: 10.4018/978-1-59904-849-9.ch089
OnDemand PDF Download:
No Current Special Offers


The importance of juice beverages in daily food habits makes juice authentication an important issue, for example, to avoid fraudulent practices. A successful classification model should address two important cornerstones of the quality control of juicebased beverages: to monitor the amount of juice and to monitor the amount (and nature) of other substances added to the beverages. Particularly, sugar addition is a common and simple adulteration, though difficult to characterize. Other adulteration methods, either alone or combined, include addition of water, pulp wash, cheaper juices, colorants, and other undeclared additives (intended to mimic the compositional profiles of pure juices) (Saavedra, García, & Barbas, 2000).
Chapter Preview

Genetic Algorithms

A GA (Holland, 1975)(Goldberg, 1989) is a recurrent and stochastic process that operates with a group of potential solutions to a problem, known as genetic population, based on one of the Darwin’s principles: the survival of the best individuals (Darwin, 1859).

Key Terms in this Chapter

Spectroscopy (Spectrometry): Production, measurement and analysis of electromagnetic spectra produced as a result of the interactions between electromagnetic radiation and matter, such as emission or absorption of energy.

Absorbance: Function (usually logarithmic) of the percentage of transmission of a wavelength of light through a liquid.

Variable Selection: Selection of a subset of relevant variables (features) which can describe a set of data.

Artificial Neural Network: Interconnected group of artificial neurons that uses a mathematical or computational model for information processing. They are based on the function of biologic neurons. It involves a group of simple processing elements (the neurons) which can exhibit complex global behaviour, as result of the connections between the neurons.

Spectrum: Intensity of a electromagnetic radiation across a range of wavelengths. It represents the intensity of emitted or transmitted energy versus the energy of the received light.

Knowledge Extraction: Explicitation of the internal knowledge of a system or set of data in a way that is easily interpretable by the user.

Evolutionary Technique: Technique which provides solutions for a problem guided by biological principles such as the survival of the fittest. These techniques start from a randomly generated population which evolves by means of crossover and mutation operations to provide the final solution.

Search Space: Set of all possible situations of the problem that we want to solve could ever be in. Combination of all the possible values for all the variables related with the problem.

Complete Chapter List

Search this Book: