[
https://issues.apache.org/jira/browse/MADLIB-1181?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16291776#comment-16291776
]
Rahul Iyer commented on MADLIB-1181:
------------------------------------
To add more color:
K-NN involves two major steps:
1. Find the k nearest neighbors to the required test point
2. Average the dependent variable for those k points to predict
The "average" in the step 2 is any aggregate function that computes a central
tendency of the values. For classification we use mode and for regression we
use mean as the averaging function. Both of them can be altered to incorporate
the weights - for mean we take sum(weight * value) and for mode we compute
sum(weight * 1) for each class.
> Add an option for weighted average in k-NN
> ------------------------------------------
>
> Key: MADLIB-1181
> URL: https://issues.apache.org/jira/browse/MADLIB-1181
> Project: Apache MADlib
> Issue Type: Improvement
> Components: k-NN
> Reporter: Frank McQuillan
> Assignee: Himanshu Pandey
> Priority: Minor
> Fix For: v1.14
>
>
> Follow on from
> https://issues.apache.org/jira/browse/MADLIB-1059
> (please see this JIRA for additional comments)
> MADlib does a simple average of the k-nearest neighbors to come up with the
> final value for classification and regression. Doing a weighted average
> instead
> might be a desirable functionality. The weighting for the average can be
> based on the
> distance of the k-nearest neighbors.
> We can probably provide an optional parameter to let users choose how the
> final
> score has to be computed (avg or weighted avg).
> This JIRA applies to classification and regression.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
