1 / 71

Estimating the Cluster Tree of a Density Werner Stuetzle Rebecca Nugent

This study explores the estimation of a cluster tree for a density using k-means grouping with 2 clusters to detect distinct groups in the data. The Runt Pruning algorithm is implemented to generate the cluster tree nodes.

dhoover
Télécharger la présentation

Estimating the Cluster Tree of a Density Werner Stuetzle Rebecca Nugent

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Estimating the Cluster Treeof a Density Werner StuetzleRebecca Nugent Department of StatisticsUniversity of Washington Interface / CSNA Meeting St. Louis

  2. Interface / CSNA Meeting St. Louis

  3. Interface / CSNA Meeting St. Louis

  4. Groups K-means with k = 2 Interface / CSNA Meeting St. Louis

  5. Interface / CSNA Meeting St. Louis

  6. Detect that there are 5 or 6 distinct groups. • Assign group labels to observations. Interface / CSNA Meeting St. Louis

  7. Interface / CSNA Meeting St. Louis

  8. Interface / CSNA Meeting St. Louis

  9. Interface / CSNA Meeting St. Louis

  10. Interface / CSNA Meeting St. Louis

  11. Interface / CSNA Meeting St. Louis

  12. Interface / CSNA Meeting St. Louis

  13. Interface / CSNA Meeting St. Louis

  14. Interface / CSNA Meeting St. Louis

  15. Interface / CSNA Meeting St. Louis

  16. Interface / CSNA Meeting St. Louis

  17. Interface / CSNA Meeting St. Louis

  18. Interface / CSNA Meeting St. Louis

  19. Interface / CSNA Meeting St. Louis

  20. rs = 2 rs = 1 rs = 5 Interface / CSNA Meeting St. Louis

  21. Interface / CSNA Meeting St. Louis

  22. Note large tree fragments in bottom left panel Interface / CSNA Meeting St. Louis

  23. Interface / CSNA Meeting St. Louis

  24. Interface / CSNA Meeting St. Louis

  25. Interface / CSNA Meeting St. Louis

  26. Interface / CSNA Meeting St. Louis

  27. Interface / CSNA Meeting St. Louis

  28. Interface / CSNA Meeting St. Louis

  29. Interface / CSNA Meeting St. Louis

  30. Interface / CSNA Meeting St. Louis

  31. Interface / CSNA Meeting St. Louis

  32. Interface / CSNA Meeting St. Louis

  33. Interface / CSNA Meeting St. Louis

  34. Interface / CSNA Meeting St. Louis

  35. Interface / CSNA Meeting St. Louis

  36. Interface / CSNA Meeting St. Louis

  37. Interface / CSNA Meeting St. Louis

  38. Interface / CSNA Meeting St. Louis

  39. Interface / CSNA Meeting St. Louis

  40. Interface / CSNA Meeting St. Louis

  41. Interface / CSNA Meeting St. Louis

  42. Interface / CSNA Meeting St. Louis

  43. Interface / CSNA Meeting St. Louis

  44. Interface / CSNA Meeting St. Louis

  45. Interface / CSNA Meeting St. Louis

  46. Interface / CSNA Meeting St. Louis

  47. Interface / CSNA Meeting St. Louis

  48. rs = 2 rs = 1 rs = 5 Interface / CSNA Meeting St. Louis

  49. rs = 2 rs = 1 rs = 5 Runt Pruning algorithm: generate_cluster_tree_node (mst, runt_size_threshold) { node = new_cluster_tree_node node.leftson = node.rightson = NULL node.obs = leaves (mst) cut_edge = longest_edge_with_large_runt_size (mst, runt_size_threshold) if (cut_edge) { node.leftson = generate_cluster_tree_node (left_subtree (cut_edge), runt_size_threshold) node.rightson = generate_cluster_tree_node (right_subtree (cut_edge), runt_size_threshold) } return (node)} Interface / CSNA Meeting St. Louis

  50. Interface / CSNA Meeting St. Louis

More Related