Transformational Derivation of Greedy Network Algorithms from Descriptive Specifications

  • Authors:

  • Juan Eduardo Durán

  • Affiliations:

  • -

  • Venue:
  • MPC '02 Proceedings of the 6th International Conference on Mathematics of Program Construction
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this work an approach for the transformational development of efficient imperative network algorithms is presented which is based on M枚ller's algebra of formal languages. We use a very flexible methodology that contemplates the description of a rather general derivation method, the use of different derivation procedures to support the phases of this method, and the derivation of algorithm schemes based on such procedures. We propose a method that consists of the formulation of a descriptive specification, the development of an abstract recursive algorithm, and the derivation of an efficient imperative algorithm. As an example, we present algorithm schemes and derivation procedures for the development of efficient greedy algorithms, which are illustrated with the derivation of a shortest path arborescence algorithm.