Knowledge reduction by combining interval Type-2 Fuzzy similarity measures and interval Type-2 Fuzzy formal lattice

Knowledge and concepts play a crucial role in human language processing. Human behavior varies in different environments, making it difficult to convey uniform definitions of these abstract notions. Furthermore, human language’s ambiguity can lead to different interpretations of a concept depending on the context. In knowledge representation, multiple context implications often increase time complexity and cause misunderstanding. To reduce and represent knowledge in concepts, we propose a hybrid technique that involves calculating concept similarity, estimating typicality, identifying concept prototypes and constructing the Interval Type-2 Fuzzy Formal Lattice (IT-2F \(^2\) ${}^2$ L). Experiments were carried out using real-world data to explore concepts lexicalized in natural language. The experimental findings demonstrate that our method, which combines Interval Type-2 Fuzzy Similarity Measures and Interval Type-2 Fuzzy Formal Lattice for concepts factorization, is a powerful tool with applications in various areas of conceptual psychology. The IT-2F \(^2\) ${}^2$ L is both interpretative and objective, making it a valuable tool for analysis and decision-making.