Semantic Information and Artificial Intelligence

Abstract

For a computational system to be intelligent, it should be able to perform, at least, basic deductions. Nonetheless, since deductions are, in some sense, equivalent to tautologies, it seems that they do not provide new information. In order to analyze this problem, the present article proposes a measure of the degree of semantic informativity of valid deductions. Concepts of coherency and relevancy, displayed in terms of insertions and deletions on databases, are used to define semantic informativity. In this way, the article shows that a solution to the problem about informativity of deductions provides a heuristic principle to improve the deductive power of computational systems.