Probabilistic Intuitionistic Fuzzy Linguistic Concept Lattice And Its Attribute Reduction Method

In the era of big data,the rapid development of artificial intelligence has greatly facilitated human production and life.Intelligent linguistic information processing has become a hot topic in uncertain environment.The linguistic terms can explicitly express human subjective preferences.But human language is extremely complex,so extended sets of the linguistic term sets are widely studied.In order to express qualitative information more comprehensively.In the uncertain environment,the fuzzy linguistic is used to describe the positive and negative evidences,and the probability is used to describe the objective randomness,which fully expresses the complex thoughts of human beings.And it is applied to concept lattice as a binary relation.The specific content of this thesis is as follows:1.In order to make the concept lattice more applicable to specific problems,the formal concepts and the partial order relations are defined based on the probabilistic intuitionistic fuzzy linguistic formal context,and then the concept lattice are constructed.By constructing the probability missing completion algorithm,the problems of high degree of ignorance,low probability and incomplete probability are solved.Based on this algorithm,the fully probabilistic intuitionistic fuzzy linguistic formal context is defined and the concept lattice is reconstructed.Finally,the algorithm is explained to be effective by comparison and analysis of examples.2.In order to solve attribute redundancy problem,attribute reduction method is studied on the probabilistic intuitionistic fuzzy linguistic formal context.Firstly,the sub-context and the partial order relation are defined based on the formal context,and the three types of attributes are analyzed.Secondly,the decision theorem of coordination set and its proof method are given by using the operator.Finally,an attribute reduction method based on the discernible matrix is proposed,and an example of investment prospect prediction is used to verify that the reduction method can reduce the complexity of the process and make the concept lattice isomorphic.3.In the specific problem,the attribute can be divided into the conditional attribute and the decision attribute.Therefore,in order to solve the attribute redundancy problem on the inconsistent probabilistic intuitionistic fuzzy linguistic formal decision context.Based on particle set and knowledge granularity,this thesis defines the internal and external importance degree of the conditional attribute relative to decision attribute set,and then constructs the attribute reduction algorithm based on relative knowledge granularity on the inconsistent formal decision context.Finally,an example of teachers’ teaching evaluation is used to verify that the reduction obtained by the algorithm does not change the concept lattice structure.To sum up,the work of this thesis is to express complex human thinking,which not only realizes the description of information from both positive and negative evidences,but also expresses subjective uncertainty and objective randomness with probability.Enhance the applicability of the concept lattice to specific problems.