[
https://issues.apache.org/jira/browse/SOLR-11943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16607107#comment-16607107
]
David Smiley commented on SOLR-11943:
-------------------------------------
I appreciate that this information will be documented in the ref guide. But
this doesn't mean the _code_ should not be documented, at least a little.
Perhaps in an ideal world we'd have some fancy mechanism of making them the
same thing for API oriented things like this but alas we don't have that.
I think "LatLon...." oriented name is much clearer as to both what it contains
and in some contexts the order of the two dimensions. Consider also that we
have a field type "LatLonPointSpatialField". "Location" is vague; perhaps
someone has a circle or bounding box or neither (just a place name) for their
_location_. If you really don't agree then perhaps a 3rd person could lend
their opinion to help pick.
> Add machine learning functions for location data
> ------------------------------------------------
>
> Key: SOLR-11943
> URL: https://issues.apache.org/jira/browse/SOLR-11943
> Project: Solr
> Issue Type: New Feature
> Security Level: Public(Default Security Level. Issues are Public)
> Reporter: Joel Bernstein
> Assignee: Joel Bernstein
> Priority: Major
> Fix For: master (8.0), 7.5
>
> Attachments: SOLR-11943.patch
>
>
> This ticket will add the following functions / features:
> 1) *locationVectors* function: Reads a list of tuples that contain a
> *location* field type and returns a *matrix* of lat/long vectors.
> 2) Add support for *haversinMeters* distance measure.
> With the addition of these two functions we'll have the ability to do various
> distance based machine learning algorithms (distance matrices, clustering,
> knn regression etc...) with location data.
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
