Abstract: The rough set model algorithm based on the decision table was used in project evaluation to acquire common decision knowledge from decisions of many specialists as the criterion and the final result of project evaluation so as to improve the existing evaluation algorithm. With this algorithm, the most important factors for decision-making in a condition attribute set and the rules of data mining can be determined. In addition, it was proposed that the reduction of the condition attribute set is used as a new condition attribute set in order to reduce the complexity of decision- making and avoid the irregularity of the individual of decision-makers under the precondition of not influencing the result of decision-making.