An Approximate Nearest Neighbours Search Algorithm Based on the Extended General Spacefilling Curves Heuristic
نویسندگان
چکیده
In this paper, an algorithm for approximate nearest neighbours search in vector spaces is proposed. It is based on the Extended General Spacefilling Curves Heuristic (EGSH). Under this general scheme, a number of mappings are established between a region of a multidimensional real vector space and an interval of the real line, and then for each mapping the problem is solved in one dimension. To this end, the real values that represent the prototypes are stored in several ordered data structures (e.g. b-trees). The nearest neighbours of a test point are then efficiently searched in each structure and placed into a set of candidate neighbours. Finally, the distance from each candidate to the test point is measured in the original multidimensional space, and the nearest one(s) are chosen.
منابع مشابه
Some improvements on NN based classifiers in metric spaces
The nearest neighbour (NN) and k-nearest neighbour (k-NN) classification rules have been widely used in Pattern Recognition due to its simplicity and good behaviour. Exhaustive nearest neighbour search may become unpractical when facing large training sets, high dimensional data or expensive dissimilarity measures (distances). During the last years a lot of fast NN search algorithms have been d...
متن کاملApproximate Nearest Neighbour Search with the Fukunaga and Narendra Algorithm and Its Application to Chromosome Classification
The nearest neighbour (NN) rule is widely used in pattern recognition tasks due to its simplicity and its good behaviour. Many fast NN search algorithms have been developed during last years. However, in some classification tasks an exact NN search is too slow, and a way to quicken the search is required. To face these tasks it is possible to use approximate NN search, which usually increases e...
متن کاملGenetically Improved Presequences for Euclidean Traveling Salesman Problems
The spscefilling curve (SFC) method of Bartholdy and Platzman is an extremely fast heuristic for the Euclidean Traveling Salesman Problem. The authors show how genetic search over a parametrized family of spacefilling curves can be used to improve the quality of the the tours generated by SFC. The computational effort required grows slowly as a function of problem size, and the tours obtained d...
متن کاملOptimum Routing in the Urban Transportation Network by Integrating Genetic Meta-heuristic (GA) and Tabu Search (Ts) Algorithms
Urban transportation is one of the most important issues of urban life especially in big cities. Urban development, and subsequently the increase of routes and communications, make the role of transportation science more pronounced. The shortest path problem in a network is one of the most basic network analysis issues. In fact, finding answers to this question is necessity for higher level ana...
متن کاملMinimax rates for cost-sensitive learning on manifolds with approximate nearest neighbours
We study the approximate nearest neighbour method for cost-sensitive classification on low-dimensional manifolds embedded within a high-dimensional feature space. We determine the minimax learning rates for distributions on a smooth manifold, in a cost-sensitive setting. This generalises a classic result of Audibert and Tsybakov. Building upon recent work of Chaudhuri and Dasgupta we prove that...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1998