In this chapter, the problem of clustering is posed as a multiobjective optimization problem where some measures of fuzzy cluster validity (goodness) are optimized. Conventional Genetic Algorithms, a popular search and optimization tool, are usually used to perform the clustering, taking some cluster validity measure as the fitness function. However, there is no single validity measure that works equally well for different kinds of datasets. It may be noted that it is also extremely difficult to combine the different measures into one, since the modality for such combination may not be possible to ascertain, while in some cases the measures themselves may be incompatible.
CITATION STYLE
Maulik, U., Bandyopadhyay, S., & Mukhopadhyay, A. (2011). Multiobjective Genetic Algorithm-Based Fuzzy Clustering. In Multiobjective Genetic Algorithms for Clustering (pp. 89–121). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-16615-0_5
Mendeley helps you to discover research relevant for your work.