An efficient visualized clustering approach (VCA) for various datasets
2015
The aim of clustering is to discover the clusters based on the similarity features of objects. The present algorithm of visual access tendency (VAT) can access an exact number of clusters by its VAT image. The VAT image displays the squared shaped dark blocks along the diagonal; number of cluster information is accessed by counting the number of obtaining square blocks. Other extended versions are SpecVAT, iVAT. These procedures can access only the number of clusters for various datasets, but not producing the complete clustering results. Therefore, we extend the VAT as a complete clustering method called as visualized clustering approach (VCA) by Khun-munkres function. Our VCA can effectively access the number of clusters as well as it discovers the clustering results. Finally, we report the quality of clustering results for the proposed VCA under different versions of VAT. The quality is measured by performance measures such as clustering accuracy, normalized mutual information, and OTSU goodness; the comparative analysis and the validity of the clustering results are also reported in the experimental study.
Keywords:
- Single-linkage clustering
- Correlation clustering
- Cluster analysis
- k-medians clustering
- FLAME clustering
- Pattern recognition
- Canopy clustering algorithm
- Machine learning
- CURE data clustering algorithm
- Brown clustering
- Artificial intelligence
- Mathematics
- Computer science
- Determining the number of clusters in a data set
- Data stream clustering
- Data mining
- Fuzzy clustering
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
24
References
3
Citations
NaN
KQI