Research On Two-sided Matching Decision Method With Uncertain Language

Two-sided matching exists widely in economic management activities and daily life,such as e-commerce transaction matching,mutual selection between students and tutors,vehicle and freight matching,Person-Post matching,knowledge or technology transaction matching,marriage matching between men and women,and so on.It involves many fields of application.Two-sided matching based on multiple attributes is an important branch of Two-sided matching research.There is such a kind of Two-sided matching problem,in which each subject evaluates each other's attributes in the form of uncertain linguistic information.Uncertain linguistic information is a general term for various forms of uncertain linguistic expressions.Although there have been studies on Two-sided matching under uncertain linguistic information,they are only applicable to that with uncertain linguistic expressions in interval form.There is still a lack of relevant research on Two-sided matching problems with multi-form linguistic information and uncertain linguistic expressions with complex forms,which hinders the promotion of two-sided matching decision-making.In addition,uncertain language consisting of two linguistic phrases is an interval form of uncertain linguistic expression.It can express the uncertainties and hesitations of decision makers to a certain extent,but it can't express the degree of preference of decision makers for certain values.So,scholars have developed probabilistic uncertain language sets with uncertain language as the core.As an extension of uncertain language,probabilistic language set is a complex and uncertain linguistic expression with strong information representation ability and similarity to evidence representation in evidence theory.Therefore,this paper studies the Two-sided matching problem with uncertain language under the framework of Two-sided matching research with evidence theory as the core,and has achieved the following results:(1)On the basis of quantifying the stability of two-sided matching,a two-stage stable matching optimization model based on Two-sided matching degree and aiming at maximizing bilateral matching degree is established.(2)Considering the incompleteness of attribute weights in some two-sided matching decision-making problem under uncertain linguistic information,a weight optimization model based on maximum deviation is proposed.And the matching degree between two arbitrary subjects is solved by evidence reasoning method.Finally,a stable matching optimization model based on evidence reasoning is established.(3)For the Two-sided matching problem of which including deterministic linguistic value,uncertain linguistic value and probabilistic uncertain linguistic value,a unified expression framework of multi-linguistic information is proposed.In order to solve the consistency problem of multi-granularity information,two methods of evidential transformation under different recognition frameworks are defined.And the matching degree between two arbitrary subjects is solved by interval evidence reasoning method.Finally,a stable matching optimization model based on interval evidence reasoning is established.(4)In order to simplify the operation,a reliability calculation method of group decision information is proposed,which extends the research of this paper to the two-sided matching problem based on group decision making.