The nearest neighbor problem in an abstract metric space

被引：15

作者：

Feustel, Charles D. ^{[1
]}

Shapiro, Linda G. ^{[2
]}

机构：

[1] Virginia Polytech Inst & State Univ, Dept Math, Blacksburg, VA 24061 USA

[2] Virginia Polytech Inst & State Univ, Dept Comp Sci, Blacksburg, VA 24061 USA

来源：

PATTERN RECOGNITION LETTERS | 1982年 / 1卷 / 02期

基金：

美国国家科学基金会;

关键词：

Nearest neighbor; relational distance; relational models; metric;

D O I：

10.1016/0167-8655(82)90025-3

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Given a database of relational models, we would like to select those models that are closest to an unknown relational description. Since relational distance is expensive to compute, we want to minimize our distance calculations. In this paper we present a new method for achieving this goal in an abstract metric space.

引用

页码：125 / 128

页数：4