[SPARK-2335] k-Nearest Neighbor classification and regression for MLLib

• Type: New Feature
• Status: Resolved
• Priority: Minor
• Resolution: Duplicate
• Component/s: MLlib
• Labels: clustering features
• Assignee: Unassigned
• Reporter: Brian Gawalt
• Shepherd: Ashutosh Trivedi
• Created: 01/Jul/14 18:20
• Updated: 16/Jan/16 13:38
• Resolved: 16/Jan/16 13:38

Description

The k-Nearest Neighbor model for classification and regression problems is a simple and intuitive approach, offering a straightforward path to creating non-linear decision/estimation contours. It's downsides – high variance (sensitivity to the known training data set) and computational intensity for estimating new point labels – both play to Spark's big data strengths: lots of data mitigates data concerns; lots of workers mitigate computational latency.

We should include kNN models as options in MLLib.

Issue Links

• duplicates: SPARK-2336 Approximate k-NN Models for MLLib
• is depended upon by: SPARK-2336 Approximate k-NN Models for MLLib