A Study On Data-based Comprehensive Evaluation Method

Comprehensive evaluation is to assess objects with the description of multi-attributesstructure from a global perspective. It plays an important role in resource management,complex system optimization and other practical problems. Since the development ofcomputer and the intensive study in related fields, the methods of comprehensive evaluationhave found many applications in society, economic, technology, education, environmentand management. How to determine attribute weights and construct evaluation model aretwo critical problems in the process of comprehensive evaluation.Information system is a database with objects and attribute relations. All fields insocial and economic development are highly dependent on data, but the data informationis always incomplete. Data in database contain relations between objects and attributes.This kind of relations can be discovered by certain mathematical methods and computingequipments. The rough set theory proposed by Pawlak in1982is helpful to solve this kindof problems. By using it, the accumulated data information can be utilized to discover newknowledge from information systems, thus assisting evaluation and decision making.In order to determine the significance of attributes in comprehensive evaluation, westudy the metric of attributes correlation based on changes of knowledge induced by thatof attributes sets in decision information systems, and construct the determinationmechanism of attributes interactions, which not only reflects the significance measures,but also satisfies the structure characteristics of fuzzy measure. Further, by taking Choquetintegral as synthesis operator, we construct the comprehensive evaluation model based ondecision information system. Finally, we analyze the features and effectiveness of theproposed model with case study. The results show that the method is different essentiallycomparing with the existing ones and also has good operability and interpretability.Moreover, it can be applied in the fields such as resource allocation, performanceassessment and complex system optimization.