[SPARK-6517] Bisecting k-means clustering

• Type: New Feature
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.6.0
• Component/s: MLlib
• Labels: clustering
• Target Version/s: 1.6.0
• Assignee: Yu Ishikawa
• Reporter: Yu Ishikawa
• Shepherd: Xiangrui Meng
• Created: 25/Mar/15 00:14
• Updated: 29/Dec/16 01:24
• Resolved: 09/Nov/15 22:56

Issue Links

• blocks: SPARK-6518 Add example code and user guide for bisecting k-means
• blocks: SPARK-6519 Add spark.ml API for bisecting k-means
• blocks: SPARK-8459 Add import/export to spark.mllib bisecting k-means
• blocks: SPARK-11611 Python API for bisecting k-means
• is duplicated by: SPARK-2429 Hierarchical Implementation of KMeans
• is related to: SPARK-12974 Add Python API for spark.ml bisecting k-means
• relates to: SPARK-18821 Bisecting k-means wrapper in SparkR
• links to: [Github] Pull Request #5267 (yu-iskw)

Sub-Tasks

• Add methods to get bisecting k-means cluster structure: Resolved
• Support other distance metrics for bisecting k-means: Resolved
• Find the best `k` by cutting bisecting k-means cluster tree without recomputation: Resolved