What question did this study set out to answer?

The aim is to evaluate how six types of error perturbation affect the performance of various clustering algorithms.

September 1, 1980

An Examination of the Effect of Six Types of Error Perturbation on Fifteen Clustering Algorithms

Puntos clave

The aim is to evaluate how six types of error perturbation affect the performance of various clustering algorithms.
Artificial clusters were created with defined internal cohesion and external isolation properties.
Six error perturbations were applied to obscure the true cluster structure.
Performance evaluations were conducted on fifteen clustering algorithms, focusing on sensitivity and recovery.
Hierarchical methods displayed varying sensitivity to different types of error perturbation.
K-means algorithms showed poor recovery when using random seed points.
Alternative starting methods improved recovery performance for nonhierarchical clustering algorithms across all error types.

Resumen

An evaluation of several clustering methods was conducted. Artificial clusters which exhibited the properties of internal cohesion and external isolation were constructed. The true cluster structure was subsequently hidden by six types of error-perturbation. The results indicated that the hierarchical methods were differentially sensitive to the type of error perturbation. In addition, generally poor recovery performance was obtained when random seed points were used to start the K -means algorithms. However, two alternative starting procedures for the nonhierarchical methods produced greatly enhanced cluster recovery and were found to be robust with respect to all of the types of error examined.

Me gusta

Guardar

Me gusta

Guardar

An Examination of the Effect of Six Types of Error Perturbation on Fifteen Clustering Algorithms

Puntos clave

Resumen

Cite This Study