Yes, faceted search is my primary use case. Right now we are mostly modeling things like income and city as node attributes with the legacy index.
I will run some tests doing the traversal and see how it performs. -------- Clark Richey 240-252-7507 > On Jun 24, 2014, at 7:50, Michael Hunger <[email protected]> > wrote: > > This is kind of a faceted search. > > depends on the selectivity > > something like gender I would neither put in an index nor connect to a > "gender" node, both would just refer to half you data which is not selective > at all, > so I would put these as property on the node > > For others like city, income group, interests etc. I'd use "attribute"-nodes, > then either look up the attribute node from the index and check if your > current node is connected to it > or just do lookup what node your node is connected to via the "lives_in" > relationship and check properties on the end node. > > HTH, > > Michael > >> Am 24.06.2014 um 11:29 schrieb Clark Richey <[email protected]>: >> >> Ok just so that I'm 100% clear you are saying looking the first node from >> indexes and do a traversal and that it should be fast even if the query is >> actually something like find people with gender male married to a person >> with income between 40 and 60k who has a residence in the city of Columbus? >> In this case looking up people that are male will return LOTS of nodes as >> might even the second step in the traversal. >> >> Sent from my iPhone >> >>> On Jun 24, 2014, at 0:48, Michael Hunger <[email protected]> >>> wrote: >>> >>> Depending on selectivity >>> >>> Lookup by x or y from index >>> Check the other props and rels >>> You might also lookup columbus and check >>> From node to columbus >>> >>> It is a local traversal then should be fast >>> >>> Sent from mobile device >>> >>>> Am 24.06.2014 um 03:51 schrieb Clark Richey <[email protected]>: >>>> >>>> All, >>>> I am writing server side extensions to neo4J. I need to be able to find >>>> nodes that match a graph pattern similar to: >>>> >>>> node has property x=foo and y= bar and outgoing relationship of type >>>> address to node with city=columbus. That could even extend another node >>>> out. I know I can find one end of the above relationship and then filter >>>> but this doesn’t work well when the pattern I want to match is more like >>>> node a -> node b -> node c with various property lookups on each node. >>>> >>>> This is also a large graph so starting at any given node could easily >>>> yield a million nodes from which to start a traversal. >>>> >>>> Thanks for the help! >>>> >>>> -------------------- >>>> Clark Richey, CTO >>>> FactGem >>>> 240-252-7507 >>>> [email protected] >>>> >>>> >>>> >>>> >>>> Need immediate assistance? Please try: >>>> >>>> Cate Downing, Assistant Beth Price, Assistant >>>> [email protected] [email protected] >>>> 614.325.2404 614.365.0740 >>>> >>>> This message and any included attachments are property of FactGem and its >>>> affiliates, and are intended only for the addressee(s). The information >>>> contained herein may include trade secrets or privileged or otherwise >>>> confidential information. Unauthorized review, forwarding, printing, >>>> copying, distributing, or using such information is strictly prohibited >>>> and may be unlawful. If you received this message in error, or have reason >>>> to believe you are not authorized to receive it, please promptly delete >>>> this message and notify the sender by e-mail. Thank you. >>>> >>>> -- >>>> You received this message because you are subscribed to the Google Groups >>>> "Neo4j" group. >>>> To unsubscribe from this group and stop receiving emails from it, send an >>>> email to [email protected]. >>>> For more options, visit https://groups.google.com/d/optout. >>>> <DB3DED71-9092-4039-BC47-7254B572F9D5> >>> >>> -- >>> You received this message because you are subscribed to the Google Groups >>> "Neo4j" group. >>> To unsubscribe from this group and stop receiving emails from it, send an >>> email to [email protected]. >>> For more options, visit https://groups.google.com/d/optout. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Neo4j" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> For more options, visit https://groups.google.com/d/optout. > > -- > You received this message because you are subscribed to the Google Groups > "Neo4j" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
