You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by Joe Chesak <jo...@easyconnect.no> on 2010/07/16 19:58:54 UTC
Fwd: send to list
I wish to display search results on a google map. I would like to group the results such that if more than one hit is in one location, the sum of all hits at that location will show up on an icon.
By one location, I mean one square on a grid of 10 x 10 squares = 100 squares. I am trying out facets for this, but the request string is enormous, and when I send a request for 30 facets the query time increases from 5ms to 355ms.
Here is what such a query looks like...
http://theguide.srv.easyconnect.no:8080/solr/no-gul-biz-web/select?facet=true
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[25.6+TO+31]
Maybe using a grid is the wrong approach, maybe I should be thinking clustering instead of grouping. Is there a best practice for doing this?
Joe
Re: Fwd: send to list
Posted by kenf_nc <ke...@realestate.com>.
If at all possible I like to do any processing work up front and not deal
with extravagant queries. If your grid definitions don't change, or don't
change often, just assign a cell number to each 100 square grid. Then in a
pre-processing step assign the appropriate cell number to your document
along with the specific lat and lon. Then your facet query gets much
simpler.
--
View this message in context: http://lucene.472066.n3.nabble.com/Fwd-send-to-list-tp973191p973233.html
Sent from the Solr - User mailing list archive at Nabble.com.
Re: send to list
Posted by "Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>.
Hi Joe,
Take a look at the Cartesian Grid work from Patrick O'Leary here [1]. It's not fully integrated with Solr and they are moving away from it, but it'll give you a good idea of how to get started and to go about doing this...
HTH,
Chris
[1] http://www.nsshutdown.com/projects/lucene/whitepaper/locallucene_v2.html
On 7/16/10 10:58 AM, "Joe Chesak" <jo...@easyconnect.no> wrote:
I wish to display search results on a google map. I would like to group the results such that if more than one hit is in one location, the sum of all hits at that location will show up on an icon.
[cid:3362123455_49287900]
By one location, I mean one square on a grid of 10 x 10 squares = 100 squares. I am trying out facets for this, but the request string is enormous, and when I send a request for 30 facets the query time increases from 5ms to 355ms.
Here is what such a query looks like...
http://theguide.srv.easyconnect.no:8080/solr/no-gul-biz-web/select?facet=true
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[4+TO+9.4]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[9.4+TO+14.8]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[14.8+TO+20.2]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[20.2+TO+25.6]
&facet.query=lat_trie:[57.9 +TO+59.56]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[59.56 +TO+61.22]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[61.22 +TO+62.89]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[62.89 +TO+64.55]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[64.55 +TO+66.21]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[66.21 +TO+67.88]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[67.88 +TO+69.54]+AND+lon_trie:[25.6+TO+31]
&facet.query=lat_trie:[69.54 +TO+71.2]+AND+lon_trie:[25.6+TO+31]
Maybe using a grid is the wrong approach, maybe I should be thinking clustering instead of grouping. Is there a best practice for doing this?
Joe
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: Chris.Mattmann@jpl.nasa.gov
WWW: http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++