Abstract :
The dual notions of discernibility and indiscernibility play an important role in intelligent data analysis. While discernibility focuses on the differences, the indiscernibility reveals the similarities. By considering them together in a same framework, one is able to obtain new insight of data. The main objective of the paper is to apply discernibility and indiscernibility to conflict analysis, a theory dealing with opinions of a set of agents on a set of issues. In particular, we are interested in the problem of issue reduction, so that a reduced set of issues can be obtained without loss of crucial information of the original set of issues. Extending the results from rough set theory, three types of issue reducts are introduced. They correspond to discernibility, indiscernibility, and discernibility-and-indiscernibility reducts, respectively. The results of this paper may offer a new research direction in rough set analysis in general, and conflict analysis in particular.
Keywords :
rough set theory; conflict analysis; discernibility framework; indiscernibility framework; intelligent data analysis; issue reduction; rough set theory; Atmosphere; Competitive intelligence; Computational intelligence; Computer science; Data analysis; Data mining; Information systems; Machine learning; Rough sets; Set theory;
