Fuzzy-Classical Linguistic Concept Acquisition Approach Based on Attribute Topology

Kuo Pang, Ning Kang, Li Zou, Mingyu Lu

Abstract: There is a large amount of fuzzy linguistic-valued data in real life. For the complexity of fuzzy linguistic concept acquisition, in this paper, we propose a Fuzzy-classical Linguistic Concept Lattice Acquisition (FLCLC) approach based on attribute topology. Specifically, the fuzzy-classical linguistic concept induced operator is proposed according to the fuzzy linguistic formal context. On this basis, the fuzzy linguistic attribute topology is generated by the fuzzy linguistic coupling relationship between attributes. Furthermore, in order to improve the interpretability of the fuzzy-classical linguistic concept acquisition process, we traverse the weight paths of the coarse fuzzy linguistic attribute topology, and the set of attributes obtained by traversing the paths is the intents of the fuzzy-classical linguistic concept, and the weights on the paths are the extents of the fuzzy-classical linguistic concept, resulting in the fuzzy-classical linguistic concept. Finally, examples are used to demonstrate the effectiveness and practicality of our proposed approach.