> > Apartments would belong to a polygon district ...
You might wanna use this AQL function, but be sure to combine it with a geo-index accelerated filter (I guess the min and max values of x and y of all polygon points for a bounding box?): https://docs.arangodb.com/3.2/AQL/Functions/Geo.html#isinpolygon ... or just a simple tag/category. That would be better performance-wise, as it could use a hash (array) index. Something to consider: if you wanted to search for a user-specified district, but also include neighboring districts, then a graph to store district relations could come in handy. <https://lh3.googleusercontent.com/-fYz8gvG8Rbw/WiFb1xyXLYI/AAAAAAAAADg/4sZpu90CkSsUE-HYo1e_Ldv1V4VjoA2bgCLcBGAs/s1600/graph-district-map.png> Of course, you need to have this polygon data and the relationships. A simple radius based search does not require any extra data. If you track things like views, matches or how many types of apartments/features there are, then there are probably many ways to calculate an edge weight based on these numbers. But you could also store some kind of ranking based on that per apartment offer and use it an a document-style rather than a graphy query. -- You received this message because you are subscribed to the Google Groups "ArangoDB" 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.
