08. Comparing DBSCAN and k-means Clustering

You would like to cluster your data into a fixed number of clusters

Performance may vary based on number of iterations

There are an unknown number of clusters present in your data but you know something about the characteristics you expect clusters to have (i.e. their density in parameter space).

The data contain outliers that you want to exclude from clusters.

No need to define a termination / convergence criteria for this algorithm.

There are an unknown number of clusters present in your data but you know something about the characteristics you expect clusters to have (i.e. their density in parameter space).

The data contain outliers that you want to exclude from clusters.

No need to define a termination / convergence criteria for this algorithm.

There are an unknown number of clusters present in your data but you know something about the characteristics you expect clusters to have (i.e. their density in parameter space).

The data contain outliers that you want to exclude from clusters.

No need to define a termination / convergence criteria for this algorithm.

You would like to cluster your data into a fixed number of clusters

Performance may vary based on number of iterations

You would like to cluster your data into a fixed number of clusters

Performance may vary based on number of iterations

There are an unknown number of clusters present in your data but you know something about the characteristics you expect clusters to have (i.e. their density in parameter space).

The data contain outliers that you want to exclude from clusters.

No need to define a termination / convergence criteria for this algorithm.