Databases
StackOverflow2013
Postsdbo
PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
    primarykey
    data
    text
    <p>I have a large number of 3d points paired with importance factors.</p> <p>Each user has a six points. For example: Person Charlie has 6 points: (22,44,55) is his first point with an importance factor of 3, (10,0,0) is his second vector with a importance factor of 2.8 all the way down to his sixth point that is (100,300,200) with an importance factor of 0.4. </p> <p>What I would like to do is to find the person that is most similar to Charlie without iterating through every single other person. Essentially minimizing this function for every user (ie, matching up the right six points from that user to Charlie):</p> <pre><code>pythagoras(point, point2) * max(importance_factor, importance_factor2) * (abs(importance_factor - importance_factor2) + 1) </code></pre> <p>And then finding the user that is most similar to Charlie over all by choosing the one with the lowest cost. I've written the code the dumb way for now (by doing lots of loops) but I'm looking for a way to properly handle the fact that there are multiple points AND importance factors.</p> <p>I started looking into spacial indexes, but I don't think they will work since I have multiple points, but perhaps I could unroll the points into a higher dimensional point? So instead of 6 points in 3 dimensions I could have 1 point in 18 dimensions? Still can't handle the importance factor, but it would better than nothing.</p> <p>Unfortunately I can't us vectors and cosines here, since (1,1,1) and (400,400,400) are <em>very</em> opposite things.</p> <p>Any ideas?</p>
    singulars
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PTQuestion
    1. USzachaysan
    1. USzachaysan
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
      1. USottomeister
      1. VTFavorite
    1. COI'm not an expert at algorithms, but would it mean that you need to put some kind of priority on either the distance or the weight to narrow down your data set? Like saying, first consider the N closest and then sort the weight? To me it seems like in order to consider that function, you would have to test every combination.
      singulars
      1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
      1. USjdi
    2. COYou can calculate the euclidean distance from this point to other points. Is the "importance factor" a separate dimension, or just a weight on the existing vectors?
      singulars
      1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
      1. USJoel Cornett
    3. COJust a weight on the existing points.
      singulars
      1. PONeed a proper data structure or index for fast user lookup based on 3d points and importance factors
      1. USzachaysan
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload