New Strategies for Evaluating the Performance of Typical Testor Algorithms

Abstract

Abstract. Typical testors have been used in feature selection and supervised classification problems in the logical combinatorial pattern recognition. Several algorithms have been used to find the set of all typical testors of a basic matrix. This algorithms are based on different heuristics. There is no doubt these algorithms find the set of all typical testors. However, the time spent on this search strategies, differs between them. Due to size of this set, the search time is a critical factor. There is not a standard procedure to evaluate the time performance of typical testors algorithms. In this paper we introduce a strategy to solve this problem through a new set of test matrices. These test matrices have the property that the set's cardinality of all typical testors is known in advance.

A Theoretical and Practical Framework for Assessing the Computational Behavior of Typical Testor-Finding Algorithms

“…Let A = [a ij ] m×n and B = [b ij ] m ×n be two basic matrices. In [2] the operators φ and θ were defined on pairs of matrices A and B. The result of a ϕ operation is a new Boolean matrix obtained by concatenating two basic matrices with the same number of rows.…”

“…However, a formal and convenient strategy for selecting matrices for algorithm testing is certainly viable. In [2], a feasible strategy for studying the behavior of the TTAs was presented for the first time. That strategy was based on the construction of test matrices (TM) which are basic matrices whose sets of typical testors can be determined in advance.…”