Off-line-access to remote databases with fuzzy searching elements

The author presents some approaches for the development of a wide area distributed databases system with remote off-line access. On the basis of such approaches different systems of commercial information processing may be constructed. He attempts to overcome the negative moments by comparing the texts of commercial offers and calculation criteria of its “likeness”. An uncertainty, obtained after such comparison, is analyzed and handled by a special algorithm. He also describes an algorithm of text identity comparison based on the fuzzy set theory. He then offers a new “hybrid” algorithm that tries to minimize the negative moments. The main goals of the analysis are: selection of “like” text descriptions on the basis of the criteria of identity; independence from new items in some classifiers or database field lists; and an automatic search of corresponding text descriptions of database, database fields, or data