What if i want to do the K-NN request without getting result in a time?

[Henryypc]

Hello, davidmoten
Thanks for reading my questions.
I want to do the K-NN query to a MBR but i don't know how much will k be. I want to get the result set incrementally. For example.It's a while loop. In each round, i want to get 5 more K-NN MBRs. What can i do to reach this goal in your code?

Looking forward to your reply.
Thanks very much!

BTW, your code really helped me a lot! Thank you very much!

[davidmoten]

What determines when k is enough for you?

[Henryypc]

Thank you very much for reply.
The terminating condition is depended by the data i get from Rtree, so i don't know when the k is enough. I want to design a dynamic algorithm to search rtree by an increasing radius r. And in each round,r=r+lambda, i don't need to search the MBR within radius r in each round, i only need to search the MBR between r and r+lambda. So I plan to mark the distance in the each round and continue in the next round. But i don't know how to...
Thank you again!

[davidmoten]

Interesting question. Essentially you want to be able to search with a custom geometry which is a donut.

In the README there are a list of conditions that need to be met for an RTree to be an applicable data structure for a custom geometry. They are:

• distance(r) >= 0 for all rectangles r
• if rectangle r1 contains r2 then distance(r1)<=distance(r2)
• distance(r) = 0 if and only if the geometry intersects the rectangle r

The trouble is for a donut the second condition does not always hold. Imagine a donut inside a rectangle r1 and another rectangle r2 inside the hole of the donut. The distance from the donut to r1 is 0 yet the distance from the donut to r2 is > 0.

Given this limitation I think you will need to do a search of r+lambda and remove the results of a search of r.

[Henryypc]

I think i understand what you mean. I will try what you said. Thank you for answering my queations. Thank you a lot!

[Henryypc]

Thank you for your help, and I find the algorithm i want to use, which is given in this paper.

Distance Browsing in Spatial Databases
GÍSLI R. HJALTASON and HANAN SAMET
University of Maryland

Maybe it can help you for your R-tree related project.

[davidmoten]

Cool. I'll have a look.

On 19 June 2016 at 16:28, Henryypc notifications@github.com wrote:

Thank you for your help, and I find the algorithm i want to use, which is
given in this paper.

Distance Browsing in Spatial Databases
GÍSLI R. HJALTASON and HANAN SAMET
University of Maryland

Maybe it can help you for your R-tree related project.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#52 (comment),
or mute the thread
https://github.com/notifications/unsubscribe/AATa650zUzAALiEbIHNj0a8dwP0rWu22ks5qNOGXgaJpZM4I0FY6
.

[davidmoten]

That paper is great. I'll put that on my backlog to consider integrating with this project. Thanks for the info!

[Henryypc]

That's cool！