This paper elaborates on the improvement of an evolutionary algorithm for clustering (EAC) introduced in previous work. Four new features are proposed and empirically assessed in seven datasets, using two fitness functions. Statistical analyses allow concluding that two proposed features lead to significant improvements on the original EAC. Such features have been incorporated into the EAC, resulting in a more computationally efficient algorithm called F-EAC (Fast EAC). We describe as an additional contribution a methodology for evaluating evolutionary algorithms for clustering in such a way that the influence of the fitness function is lessened in the assessment process, what yields analyses specially focused on the evolutionary operators.

This paper tackles the problem of showing that evolutionary algorithms for fuzzy clustering can be more efficient than systematic (i.e. repetitive) approaches when the number of clusters in a data set is unknown. To do so, a fuzzy version of an Evolutionary Algorithm for Clustering (EAC) is introduced. A fuzzy cluster validity criterion and a fuzzy local search algorithm are used instead of their hard counterparts employed by EAC. Theoretical complexity analyses for both the systematic and evolutionary algorithms under interest are provided. Examples with computational experiments and statistical analyses are also presented.

Structural health monitoring is based on the development of reliable and robust indicators able to detect, locate, quantify or even predict damage. Studies related to damage detection in civil engineering structures are of interest to researches in this area. Indeed, the detection of structural changes likely to become critical can prevent the occurrence of major dysfunction associated with social, economic and environmental consequences. Recently, many researchers have focused on dynamic assessment as part of structural diagnosis. Most of the studied techniques are based on time or frequency domain analyses to extract information from modal characteristics or based on indicators built from those parameters. The main goal of this study relies on the application of symbolic data analysis coupled with classification methods to detect structural damage, especially using raw data (i.e. in situ measurements). Modal parameters, such as natural frequencies and mode shapes, are also considered in the analysis. In order to attest to the efficiency of the proposed approach, experimental investigations in the laboratory and on two real case studies – railway and motorway bridges – are carried out. It is shown that symbolic data analysis coupled with classification methods is able to distinguish structural conditions with very encouraging results.

No abstract

No abstract