Abstract: Attribute reduction is one of the key problems of the rough set theory. In order to effectively use the rough set theory to deal with the problem of attribute reduction in Decision Tables containing Continuous-Valued Attributes (DTCVA) directly,a new variable precision rough set model and a heuristic algorithm for attributes reduction in DTCVA were developed. Simulation results show that the proposed approach is effective for reduction of continuous-valued attributes, and more efficient than the classical rough set approaches in processing attribute reduction in decision information systems containing continuous-valued attributes.

Key words: Rough set, Attribute reduction, Variable precision rough set, Attributes significance