Generating Optimum Number of Clusters Using Median Search and Projection Algorithms
| Autor: | Rajappa Veluru, L Suresh, Jay B. Simha |
|---|---|
| Rok vydání: | 2010 |
| Předmět: |
SQL
k-medoids Computer science Correlation clustering Single-linkage clustering k-means clustering Data structure computer.software_genre Column (database) Complete-linkage clustering Data warehouse Determining the number of clusters in a data set ComputingMethodologies_PATTERNRECOGNITION CURE data clustering algorithm Convergence (routing) Cluster (physics) Affinity propagation Algorithm design Data mining Cluster analysis computer Computer Science::Databases k-medians clustering computer.programming_language |
| Zdroj: | AINA Workshops |
| DOI: | 10.1109/waina.2010.196 |
| Popis: | K-means Clustering is an important algorithm for identifying the structure in data. Kmeans is the simplest clustering algorithm. This algorithm takes a predefined number of clusters as input. Mean stands for an average, an average location of all the members of a particular cluster. This algorithm is based on random selection of cluster centers and iteratively improving the results. In this work, a novel approach to seeding the clusters with the latent data structure is proposed. This is expected to minimize: The need for number of clusters apriory Time for convergence by providing near optimal cluster centers. Also these algorithms are tested on the latest standards for data warehouses -- the column store databases. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|