Resultado de búsqueda
A vantage-point tree (or VP tree) is a metric tree that segregates data in a metric space by choosing a position in the space (the "vantage point") and partitioning the data points into two parts: those points that are nearer to the vantage point than a threshold, and those points that are not.
VP-Tree (Vantage Point Tree) is a data structure that enables efficient nearest neighbor search in metric spaces, with applications in machine learning, computer vision, and information retrieval.
Un VPT es un árbol binario en el que cada nodo representa un subconjunto S de individuos del conjunto inicial, utiliza un elemento especial del conjunto llamado pivote (vantage point) para dividir el conjunto S en dos subconjuntos, uno por cada hijo.
Vantage Point Tree (or vp tree) is a space partitioning data structure that allows for efficient querying of nearest neighbors in high dimensional spaces.
A vantage point tree allows to do fast but exact nearest neighbor searches in any space provided that you have a distance function to measure the distance between any two points in that space. This implementation might need some tweaks in case it is used to index a very large number of points (especially the select_vp function in the code).
Vantage Point Tree: Construction 1. Select a vantage point v in X (eg. following a uniform distribution); 2. Compute the distances d(v, xi) between v and each point xi in X; 3. Take the median μ of these distances; 4. Divide X in 2 sets using μ as a threshold: a. X left the set of points closest to v is put at the left b. X right
Distance-based methods that use the LSH to define the distance metric between points, and use indexing structures such as Vantage-Point-Trees (VPT) [25, 26] to do fast nearest neighbor search (such as HAC-T [18]). Collision-based methods are ideal when the LSH buckets have a good fit with the desired clusters.
