Modelling multi-agent three-way decisions with decision-theoretic rough sets. (English) Zbl 1248.68505

Yang, Xiaoping; Yao, Jingtao

Fundam. Inform. 115, No. 2-3, 157-171 (2012).

Summary: The decision-theoretic rough set (DTRS) model considers costs associated with actions of classifying an equivalence class into a particular region. With DTRS, one may make informative decisions in the form of three-way decisions. Current research mainly focuses on single agent DTRS which is too complex for making a decision when multiple agents are involved. We propose a multi-agent DTRS model and express it in the form of three-way decisions. The new model seeks for synthesized or consensus decisions when there are multiple decision preferences and criteria adopted by different agents. Various multi-agent DTRS models can be derived according to the conservative, aggressive and majority viewpoints based on the positive, negative and boundary regions made by each agent. These multi-agent decision regions are expressed by figures in the form of three-way decisions.

MSC:

68T42	Agent technology and artificial intelligence
68T37	Reasoning under uncertainty in the context of artificial intelligence

Keywords:

decision-theoretic rough sets; three-way decisions; probabilistic rough sets; multi-agent decision makings