Ana María Badiola Sánchez

A taxonomy for the crossover operator for real-coded genetic algorithms: An experimental study

The main real‐coded genetic algorithm (RCGA) research effort has been spent on developing efficient crossover operators. This study presents a taxonomy for this operator that groups its instances in different categories according to the way they generate the genes of the offspring from the genes of the parents. The empirical study of representative crossovers of all the categories reveals concrete features that allow the crossover operator to have a positive influence on RCGA performance. They may be useful to design more effective crossover models.

Hybrid crossover operators for real-coded genetic algorithms: an experimental study

Most real-coded genetic algorithm research has focused on developing effective crossover operators, and as a result, many different types have been proposed. Some forms of crossover operators are more suitable to tackle certain problems than others, even at the different stages of the genetic process in the same problem. For this reason, techniques which combine multiple crossovers have been suggested as alternative schemes to the common practice of applying only one crossover model to all the elements in the population. Therefore, the study of the synergy produced by combining the different styles of the traversal of solution space associated with the different crossover operators is an important one. The aim is to investigate whether or not the combination of crossovers perform better than the best single crossover amongst them. In this paper we have undertaken an extensive study in which we have examined the synergetic effects among real-parameter crossover operators with different search biases. This has been done by means of hybrid real-parameter crossover operators, which generate two offspring for every pair of parents, each one with a different crossover operator. Experimental results show that synergy is possible among real-parameter crossover operators, and in addition, that it is responsible for improving performance with respect to the use of a single crossover operator.

Multiple Crossover per Couple with Selection of the Two Best Offspring: An Experimental Study with the BLX-alpha Crossover Operator for Real-Coded Genetic Algorithms

In this paper, we propose a technique for the application of the crossover operator that generates multiple descendants from two parents and selects the two best offspring to replace the parents in the new population. In order to study the proposal, we present different instances based on the BLX-? crossover operator for real-coded genetic algorithms. In particular, we investigate the influence of the number of generated descendants in this operator, the number of evaluations, and the value for the parameter ?. Analyzing the experimentation that we have carried out, we can observe that it is possible, with multiple descendants, to achieve a suitable balance between the explorative properties associated with BLX-? and the high selective pressure associated to the selection of the two best descendants.

A Prediction System for Cardiovascularity Diseases Using Genetic Fuzzy Rule-Based Systems

In this paper we present a fuzzy rule-based system to predict cardiovascularity diseases. The input variables of the system are the most influent factors for that type of diseases and the output is a risk prediction of suffering from them. Our objective is to get an accurate prediction value and a system description with a high degree of interpretability. We use a set of examples and a design process based on genetic algorithms to obtain the components of the fuzzy rule-based system.

Parameterization of Structural Faults in Large Historical Constructions for Further Structural Modelling Thanks to Laser Scanning Technology and Computer Vision Algorithms

Laser scanning technology has evolved significantly in the last decade, particularly, in those applications in terrestrial environments dealing with the documentation and inspection of civil engineering and architectural constructions. Even though there exist mature procedures to convert the so-called LiDAR point clouds in CAD models or even FEM models, the current trends in the technology are related to the automation of these operations. The development of robust automatic procedures for data segmentation and interpretation it is a key aspect so that the technology can definitely be accepted as a basic, accurate, and robust tool for reverse engineering of existing constructions. This paper presents the application of laser scanning technology to the structural evaluation of the Medieval Wall of Guimaraes (Portugal). This laser scanning survey was conducted with the aim of having an accurate and detailed geometrical model of the large masonry construction that includes the existing deformations and structural faults. The parameterization of structural damages was possible thanks to the highly detailed point cloud collected, and its processing using computer vision algorithms. The geometric models obtained could be used for further structural analysis of the entire wall.