|
Optimizing Multiple Objectives for Clustering| old_uid | 17829 |
|---|
| title | Optimizing Multiple Objectives for Clustering |
|---|
| start_date | 2019/10/15 |
|---|
| schedule | 10h |
|---|
| online | no |
|---|
| summary | Clustering is an unsupervised exploratory data analysis tool which groups thedata on the basis of some similarity/dissimilarity metric such that a predefinedcriterion is optimized. The problem of clustering is therefore essentially oneof optimization. The use of metaheuristics like genetic algorithms has been madesuccessfully in the past for clustering a data set. It is to be noted that theclustering problem admits a number of criteria or cluster validity indices thathave to be simultaneously optimized for obtaining improved results. Hence inrecent times the problem has been posed in a multiobjective optimization (MOO)framework and popular metaheuristics for multiobjective optimization have beenapplied. In this talk, we will first briefly discuss about the fuzzy c-meansalgorithm and the basic principles of MOO. Subsequently it will be shown how apopular multiobjective optimization algorithm may be used for solving theclustering problem. Since such algorithms provide a number of solutions, a wayof combining the multiple clustering solutions so obtained into a single oneusing supervised learning will be explained. The talk will conclude bydemonstrating an application of the multiobjective clustering technique on geneexpression data sets. |
|---|
| responsibles | Piwowarski |
|---|
| |
|