Computing Reviews

Non-partitionable point sets
Avis D. Information Processing Letters19(3):125-129,1984.Type:Article
Date Reviewed: 07/01/85

Records in a database are often considered as points in a space of dimension equal to the number of data fields. Then the design of certain data structures leads to questions about partitions of the space. A set S of n points in d-space is said to be &agr;(n)-partitionable if there exists a set of d mutually intersecting hyperplanes that devide d-space into 2d open regions such that no 2d − 1 regions contain more than &agr;(n) points of S. Every planar set S is 3-4- n-partitionable and every set in 3-space is 23-24- n-partitionable; this was proved by Willard [1] and Yao [2], respectively. In this paper, it is shown that there exist sets S of arbitrary cardinality in d-space, d ≥5, for which d2 + 1 open regions together contain at least nd2 points of S, in any partition by d intersecting hyperplanes. Moreover, at least 2dd2 − 1 open regions contain no points of S. Thus, the effective balanced-tree approaches of the cited authors may not be generalized to dimensions d ≥5.


1)

Willard, D. E.Polygon retrieval, SIAM J. Comput. 11 (1982), 149–165.


2)

Yao, F.09MA 3-space partition and its applications (Extended Abstract), Proc. of the 15th annual ACM symp. on theory of computing (Boston, MA, April 25-27, 1983), ACM, New York, 1983, 258–263.

Reviewer:  W. C. Rheinboldt Review #: CR109126

Reproduction in whole or in part without permission is prohibited.   Copyright 2024 ComputingReviews.com™
Terms of Use
| Privacy Policy