Well-explained indexing and search algorithms in metric spaces

Question

I need to implement some kind of metric space search in Postgres (*) (PL or PL / Python). So, I am looking for good sources (or documents) with a very clear and concise explanation of the mechanism behind these ideas, so that I can implement it myself.

I would prefer clarity on efficiency.

(*) The need for this is described better here .

Answer 1

PostGIS, , - . , , GiST.

, , - . , Navarro, Baeza-Yates, Sutinen Tarhio, IEEE Data Engineering Bulletin, 2001; Citeseer . Locality Sensitive Hashing - , , .

Answer 2

BK-Trees , , . - . .

, Postgres . , GIST, , , . - , , , , , .

Answer 3

http://sisap.org, , BK-. C, .

Answer 4

Some methods for finding space that might help you are Hill-Climbing, Neural Network Training, Genetic Algorithm, and Particle Swarm.

You also need to determine the distance metric from the metric space. You did it? (& out of curiosity what it is if you did it)