Business Intelligence Investigate

Hybrid Algorithm for Clustering Mixed Data Sets

Download now Free registration required

Executive Summary

Clustering is one of the data mining techniques used to group similar objects into different meaningful classes known as clusters. Objects in each cluster have maximum similarity while the objects across the clusters have minimum or no similarity. This kind of partitioning of objects into various groups has many real time applications such as pattern recognition, machine learning and so on. In this paper, the authors review a clustering algorithm based on genetic K-means and compare it with GKMODE and IGKA. The algorithm works well for both numeric and discrete values.

  • Format: PDF
  • Size: 218.75 KB