Order Structure And Feature Selection Methods Of Hesitation Decision System

With the advent of the data age,all walks of life have begun to pay more attention to diverse data information processing.As a set-valued decision information system,the hesitation value decision information system can more objectively reflect the uncertainty information contained in the data due to the hesitation and ambiguity of its attribute values.The classic Pawlak rough set theory is a theory that can effectively process data and information in complex systems.However,the classic rough set model is based on the equivalence relationship for data and information mining,and the processing of complex systems such as hesitation value decision systems is often limited.Therefore,this article mainly promotes the classical rough set model for the data of hesitation fuzzy value type,and proposes the rough set model under the order relation and the fuzzy rough set model under the fuzzy order relation,and respectively solves the uncertainty of the model.The characteristics are mined,and finally the feature selection result of the value decision information system is obtained.In the research of hesitation value decision information system,an important research topic is to dig out the special relationship between different samples from multiple information under the attributes reflected by hesitation value.Due to the shortcomings in the current research on the order structure of hesitation fuzzy elements,chapter 3 of this paper constructs a new dominance to describe the order relationship between samples in the hesitation value decision information system.Combined with the classical rough set theory proposed by Pawlak,the uncertainty measure is further constructed by using the aforementioned superiority degree to measure the ability of the attribute subset to maintain the partial order of decision-making.Based on this uncertainty measure,a forward greedy feature selection method is proposed.At the end of this chapter,combined with examples,the feature selection process of the hesitation value decision information system under the order relation is given.In order to further study the hesitation value decision information system,this article proposes a new fuzzy order relationship in Chapter 4 to describe the fuzzy order relationship of samples in the hesitation value decision information system.Combined the classic fuzzy rough set theory and fuzzy precision,an entropy model for measuring uncertainty is constructed.Using this model,the forward greedy feature selection method is used,combined with examples,to give the hesitation value decision under the fuzzy order relationship the feature selection process of the information system confirms the validity and rationality of the model proposed in this chapter.The research ideas and results of this article have referent significance for measuring partial order relations of hesitant fuzzy systems and constructing uncertainty measures.