| Summary: | For decision makers, the data property has a direct influence on the selection of decision making approaches and the reliability of decision results. Because of the complexity and diversity of practical decision data, some traditional decision approaches are not very good at reflecting the actual problem. For this, we propose a decision making model in the environment of complex structure data. The aim of this model is to discover the underlying community structure of the data by taking all aspects of original information into account. In this paper, the considered data is diversity, not only in structure but also in representation. What is more, a missing data compensation method is proposed by considering the information losing situation in practical decision making problem. Research shows that this model has great maneuverability. Especially, the proposed decision model seems more consistent with the actual decision problem, than decision model with single data structure.
|
|---|