Icon

Clustering_​And_​Elbow_​Graph

There has been no title set for this workflow's metadata.

There has been no title set for this workflow's metadata.

The workflow reads a data set to be clustered (Iris data). The workflow details how to use K-Means clustering with a number of different metrics: Inertia, Entropy, and Silhouette Coefficient. It also illustrates how to create an elbow graph.

Showing how you can loop through values of "k", calulating the distances to create an Elbow graph using KNIME Showing how you can use the same looping method as above to find optimal k. This one based on the Entropy metric. Showing another looping method you can use to find the optimal k. This one basedon Silhouette Coefficient. Value for kAppend ParameterAssess QualityExtract QualityValue for kAppend ParameterReading inIris dataset Optimized K-Means(Silhouette Coefficient) k-Means Calculate sum ofsquared errors Find Elbow Counting Loop Start Python EditVariable Loop End Variable toTable Column Line Plot Entropy Scorer Row Filter Counting Loop Start Python EditVariable k-Means Loop End Variable toTable Column Line Plot Sorter Table Reader Showing how you can loop through values of "k", calulating the distances to create an Elbow graph using KNIME Showing how you can use the same looping method as above to find optimal k. This one based on the Entropy metric. Showing another looping method you can use to find the optimal k. This one basedon Silhouette Coefficient. Value for kAppend ParameterAssess QualityExtract QualityValue for kAppend ParameterReading inIris datasetOptimized K-Means(Silhouette Coefficient) k-Means Calculate sum ofsquared errors Find Elbow Counting Loop Start Python EditVariable Loop End Variable toTable Column Line Plot Entropy Scorer Row Filter Counting Loop Start Python EditVariable k-Means Loop End Variable toTable Column Line Plot Sorter Table Reader

Nodes

Extensions

Links