Method for three-way decisions using similarity in incomplete information systems

Aiming at the problems of filling missing information and calculating conditional probability and loss function in incomplete information systems, this paper provides a novel three-way decision model based on incomplete information systems. Firstly, a new information table is obtained by filling in the missing information based on similarity, and the conditional probability calculation method is established by the idea of a TOPSIS combination with the information table. The relative loss function is calculated based on the risk avoidance coefficient under different attributes. Then, we propose the notion of interval relative loss function and give formulae to calculate the interval relative loss function values. In particular, the key steps of constructing the three-way decision model are summarized. Finally, a case study of medical diagnosis is provided to verify the validity of the proposed method. Moreover, the rationality and superiority of the presented method are verified by sensitivity analysis and comparative analysis.